Predictability of market interest rates. Panel data approach

Размещено на http://www.allbest.ru/

[Введите текст]

ФЕДЕРАЛЬНОЕ ГОСУДАРСТВЕННОЕ АВТОНОМНОЕ ОБРАЗОВАТЕЛЬНОЕ УЧРЕЖДЕНИЕ

ВЫСШЕГО ПРОФЕССИОНАЛЬНОГО ОБРАЗОВАНИЯ

«НАЦИОНАЛЬНЫЙ ИССЛЕДОВАТЕЛЬСКИЙ УНИВЕРСИТЕТ

«ВЫСШАЯ ШКОЛА ЭКОНОМИКИ»

Международный институт экономики и финансов

Бобылева Надежда Борисовна

PREDICTABILITY OF MARKET INTEREST RATES. PANEL DATA APPROACH

Выпускная квалификационная работа - БАКАЛАВРСКАЯ РАБОТА

по направлению подготовки 38.03.01 «Экономика»

образовательная программа «Программа двух дипломов по экономике НИУ ВШЭ и Лондонского университета»



Introduction

The given research is dedicated to the theme: «Predictability of market interest rates. Panel data approach». I will focus my analysis on the problem of market interest rates predictibility (particularly of the prime loan rate) in the United States of America which can be called relevant and important because of the growing interest in monetary policy decisions in general and their influence on a great number of market indicators. The financial market of the USA is one of the most important and quickly developing with a great number of instabilities and dangers. Market interest rates serve as financial indicator which influence such variables as inflation, investment decisions, exchange rates and demand on financial assets. That is why such problem as market rate predictability doesn't loose its relevance.

Only during the past several decades it has been believed that the transparency of central bank decisions is needed for the effective monetary policy. The transparency itself includes such measures as publications of data used during the process of making the final decision, detailed voting records on interest rates and conclusions made by the committee members. Such transparency is thought to be financially and economically useful and is approved by many researches due to the empirical evidence that if private sector has an opportunity to know more about Central Bank decisions, then there is an improvement in predictability of expectations.

However, such advantages of transparency were not always supported by all researchers. Several decades ago unanticipated changes in monetary policy were thought to be of higher benefit than anticipated ones. The main principal of central banks was to provide closed and non-transparent monetary policy: «a certain degree of ambiguity provides the policymaker with greater influence on the timing of monetary surprises. When there is ambiguity about policy, ha can create large positive surprises and leave the inevitable negative surprises for periods in which he is relatively more concerned about inflation».

The main aim of my research is to prove that if central bank would publish detailed voting results on market interest rates immediately after the session, this would improve the predictability of interest rates and, thus, amend the monetary policy effectiveness. Nowadays only ten Central Banks of different countries including the USA adhere to such a policy. The main conclusions of this work with the use of the example of the Federal Reserve System should show that the quick publication of a various FOMC voting measures of dissents would improve the fit of the regression for the predictors.

The practical analysis of this research would consist of two parts: using aggregated and disaggregated (panel) data. The main data useful for the work includes actual American market interest rates, forecasted interest rates for next several quarters published every month and the measure of divisibility between voting behavior of different committee members. One of the most important problems is to prove that including the dissent variable would be beneficial for the predictions of market interest rates set by the Federal Reserve System in retrospective. The hypothesis of the dissent significance would be tested with the use of such programs as Eviews and Stata.

Thus, the object of this research is the American prime loan rate, its dynamics, forecasts and the role of its' forecasts in conducting effective monetary policy.

The main sources of information for this paper data set are: Federal Reserve Economic Data (FRED - St. Louis Fed), official publications of the Federal Reserve System, the unique and most recent research results of Andrei Sirchenko and open Internet resources.



Chapter 1. Literature overview

In this part of my research I would like to present the main literature resources I used in order to conduct more explicit and deep research. All the papers presented below presume the most popular contemporary point of view that the more the public knows about the monetary policy decisions made and the country's Central Bank, the easier it would be to make forecasts and predictions about future economic and financial changes.

The first research paper which underlies the further analysis is «Central Banksґ Voting Records and Future Policy» published by Roman Horvбth, Kateшina Љmнdkovб and Jan Zбpal. The paper conducts the research where the monetary policy of real countries over a certain period of time is analyzed: the Czech Republic, Hungary, Poland, Sweden, the United Kingdom and the United States of America. The authors of the paper try to find out whether the voting of Central Bank committee members affects the future monetary policy decisions and changes. They also analyze the Committee Members' Behavior and how the whole transparency may be improved. This research arises the same problem as my research.

The second paper, which is vital for the further analysis, is «Policymakers'votes and predictability of monetary policy» by Andrei Sirchenko. Using the example of The National Bank of Poland the author explains how the publishment of voting results of Central Bank may affect the forecasting of monetary policy decisions. Thus, in that paper it is proved that the transparency of Central Banks may improve the forecasting power of macroeconomic indicators. The conclusions of the paper suspect that instead of the six-week publishment of the committee voting results, The National Bank of Poland should publish them immediately in order to reduce the information asymmetry and the uncertainty in financial markets. The author explicitly analyzes the problem of transparency and makes the same conclusions I would like to reach in my further research.

The paper of Alessandro Riboni and Francisco J. Ruge-Murcia «Dissent in Monetary Policy Decisions» helps to define the concept of the dissent of the committees of central banks of England, Sweden and the United States. The detail technique of acquiring, calculating and analyzing the dissent value is represented. The authors also explain whether the current dissents affect the voting results of future committee meetings. They note that dissents at the Federal Reserve System are less evident and frequent than the dissents in the Central Bank of England and Sweden. Using the examples of the Bank of England and the Riksbank the predictive power of monetary policy is proved.

The research of Alan S. Blinder «Monetary Policy by Committee: Why and How? » provides the readers with an explicit picture of the perfomances of committees from different countries including the United States of America. The author describes the process of the decision making about the policy decisions. He also notes the main advantages of the committee decision making over the individual decision making. Alan Blinder gives the description and recommendations about the of optimal monetary policy committee. This paper gives an opportunity to find out more about the central bank committees of different countries, their features, functions and aims.

In order to understand the specific nature of the Federal Reserve System the research «Dissents and Disagreement on the Fed's FOMC: Understanding Regional Affiliations and Limits to Transparency» of Ellen Meade would be efficient. The paper describes the vital historical changes of FOMC communication practices and their influence on the overall economic and financial tendencies. The author also estimates the dissagreements of FOMC members in a pre-1993 and post-1993 spaces, explaining the fact of such division that only in 1993 not really detailed transcripts of FOMC meetings began to be published.

One more paper serving as one of the main sources this research relies on is «Making Sense of Dissents: A History of FOMC Dissents» written by Daniel L. Thornton and David C. Wheelock. The main aim of this paper was to show the concept of the dissent using the example of the Federal Open Market Committee. The researches conducted inform the reader about the dynamics of changes in the dissent from the 1936 to the 2013 years. There are also described the main statistics about the dissent directions. Moreover, such problems as inflation and unemployment are also shown in the way of depending on the dissent values.

The research of Ellen E. Meade «The FOMC: Preferences, Voting, and Consensus» serves as an analyzing review of the FOMC in the Greenspan years. The unique data on the FOMC meeting results is collected and analyzed for the period from 1989 to 1997. The Chairman's Alex Greenspan proposals are compared with the results of other members of the meeting and the direction of this bias is determined. Moreover, the unique statistics about the change in the policymakers' opinions after the listening to the opinions of other participants is represented. What's more, this research makes the division between the official dissent (where only the voting members' dissents are calculated) and the unofficial dissent (where both the voting and non-voting members' dissents are calculated).

The further most significant paper is «Is the MPC's voting record informative about Future UK Monetary Policy?» written by Petra Gerlach-Kristen. This paper also considers whether the proper and detail information about committee meetings may influence the future monetary policy forecastibility. The research serves as a proof that the transparency of the whole macroeconomic decisions improves after the publication of MPC minutes.

With respect to the significance of the role of monetary policy in this research the paper «Transparency of Monetary Policy: Theory and Practice» written by Petra M. Geraats. The author analyzes the features of transparency and analyzes the monetary policy decisions with respect to transparency for a sample of countries. Such features of transparency as improvement of forecastability and credibility of monetary policy and reputation building are mentined and explained. The separate attention is drawn to the practical usage of transperancy using the real examples.



Chapter 2. The monetary authorities of the United States of America

2.1 The Federal Reserve System

The Federal Reserve System (the Fed) is the central bank of the United States. It is also said to be the most influential and important central bank in the world as its decisions may affect almost all the economies of different countries. Every year the Fed's influence on the world economy becomes more and more evident.

The history of The Federal Reserve System takes its roots in the crisis of 1907 when a great number of bankruptcies, financial failures and panics leaded to the creation of this institution as a way of escaping such problems. The official Federal Reserve Act about the creation of the Fed was signed by the President on the 23d of December in 1913. The main functions of the contemprorary Federal Reserve System are divided into several areas. Firstly, the Fed is responsible for the taking control of financial instabilities. Secondly, its duty is to control the actions of country's banking institutions in order to bear up the individual's interests. Thirdly, the Fed regulates a great number of variables and different macroeconomic indicators like employment and interest rate by setting its own monetary policy. Lastly, its function is to support the US depository institutions with financial services.

The Federal Reserve System's structure can be called rather complicated compared to those of other countries. The main elements are Board of Governors, twelve regional Federal Reserve Banks, Federal Advisory Council and Federal Open Market Committee (FOMC). Let's analyze these institutes more deeply.

Board of Governors

This institution consists of seven members that are chosen by the President of the USA and accepted by the Senate. Members are usually chosen from diverse Federal Reserve districts in order to escape from the biased decisions and opinions and to make the analysis of the situation more consistent. These members are also included into the Federal Open Market Committee being the majority of this Committee. The term for the Board of Governors member is fourteen years, while the term for the Board Chairman (who is also chosen by the President and accepted by the Senate) lasts four years. There exist a number of the Board of Governors' responsibilities including the participation in FOMC votings, governing commercial banks, determining the discout interest rate and making reports about economic development and perpectives of the country.

Federal Reserve Banks

Frederic Mishkin in his paper «The economics of Money, Banking, and Financial markets» properly describes this part of the Federal Reserve System: The Board of Governors' duty is to take control over the USA Federal Reserve banks, which should present their annual budgets to the Board of Governors. There exist overall twelve Federal Reserve banks in the United States and also they have their own Branches. Each Federal Reserve Bank controls one district of the USA and is responsible for the member banks located in the same area. The division of the districts and Federal Reserve Banks are shown on the Appendix 1(Federal Reserve Bulletin).

Each Bank's board consists of nine members, which also vote for the president of their regional Bank that should be cohered with the central Board of Governors. All the members are divided into three different types depending on by who they were nominated and which interests they represent. Over the half of the assets of the whole Federal Reserve System are held by New York, Chicago, and San Francisco Federal Reserve Banks.

However, the special role of the New York Federal Reserve Bank is always pointed out. It is widely known that the New York Federal Reserve Bank' stability is important for the economical safety of the whole country as they constitute the greatest part of the Federal Reserve Banks of the whole country. Moreover, the New York Bank is said to be involved in a great number of open market operations and thus to be a good source of information about foreign markets. The New York Federal Reserve Bank president is one of the most important members of the whole Federal Reserve System, as he is the only Federal Reserve President, which permanently participate in FOMC meetings. Finally, this Federal Reserve Bank is the only one of all, which is included as the member of Bank for International Settlements (BIS).

The main functions of each Federal Reserve Bank:

They provide the Federal Reserve System with the reports and information about the economic conditions of each of the district. The «beige book» which includes the reports about the state of each district is published by the Reserve Banks every two weeks before each FOMC meeting

Each Federal Reserve Bank gives advices about interest rates for the Bank's in each district

The issue of new currency and the withdraw of old currency from circulation

Each Federal Reserve Bank makes its own researches related to the monetary policy of the United States and its implications

As my thesis is mainly dedicated to the analysis of forecasted interest rate and the influence of voting results of FOMC meetings, I would like to cover the topic of the Federal Open Market Committee (FOMC) more deeply in the next section.

2.2 The Federal Open Market Committee (FOMC)

Structure and functions

The Committee consists of twelve voting members: seven are brought from the Board of Governors, while other five are the representatives of the Federal Reserve Banks including the permanent member - the president of the New York Federal Reserve Bank. Other seven representatives of Federal Reserve Banks have the right to attend the meeting and to participate in the discussions along with no right to vote. The chairman of the Board of Governors is said to be the Federal Open Market Committee chairman and the president of the New York Federal Reserve Bank is said to be vice chairman.

The law states that the FOMC meetings should be held in Washington at least four times a year. The meetings may be held more frequently depending on the current economic situation of the country. Only a limited number of individuals (including Committee members, other Reserve Bank presidents, a limited amount of staff, the Manager of the System Open Market Account) are allowed to participate or simply to visit the FOMC meeting.

The main functions of which The Federal Open Market Committee is responsible include setting monetary policy by determining the target market federal funds rate and, thus, controlling the country's money supply.

The process of decision making

In order to understand and analyze the nature of dissents among the committee members it seems really important to acquire more information about the decision making process during the FOMC meeting.

To start with, it should be mentioned that there exist three main documents significant for the FOMC meetings. The Greenbook is written by the Federal Reserve Board of Governors' Research and Statistics Division and it is constituted before every committee meeting. This document contains data set about various macroeconomic changes and indicators of a given year. It also contains a forecast for next several years. Monetary Affairs Division at the Board of Governors is responsible for the constituting of the Blue book (Monetary Policy Alternatives), where the main three ways of conducting monetary policy are suggested. As it was explained before, the Beige book also plays a vital role for the FOMC meetings. Every Reserve Bank of the country writes a detailed statement about the macroeconomic conditions and states of the district it displays. This book is published two weeks before every meeting and is the only document from the given which can be read by the public.

While describing the process of the FOMC meeting it is possible to decide the whole scenario of the meeting into four stages:

The first step begins with the staff members, which represent oral speaches about the future perspectives, current economic conditions and developments. Next after their presentations the members of the committee and the representatives of other Federal Reserve Banks share each others opinions, discuss main changes in exchange markets, interest rate, macroeconomic indicators and fiscal policy in general. It should be noted that although representatives of Reserve Banks which don't vote have no right for the final voting, they can express their opinions and may influence the voting results a lot, which is calculated in the measure of dissent during the meeting.

After such speaches and discussions the theme turns to the discussion of the monetary policy. The voting and non-voting participants exchange their opinions about the perspective ways of economy development and monetary policy.

The next step of the FOMC consists of the concrete and sequent expressment of FOMC members' opinions with arguments and stated preferences. The discussion usually begins with the verbal Chairman's policy preferences, continuing with the Board members and all of the Reserve Bank presidents.

The last step is the official voting on the policy chosen. Speaking about the federal fund interest rates, the Blue book with three different types of monetary policy is suggested to the Committee and the Chairman decides, which of the strategies he wants to stick too, thus, suggesting his own directive. Then the voting members of the Committee should vote «for» or «against» the suggested rate.

History of FOMC meetings

The conducting monetary policy decisions was inititally only the FOMC responsibility. The first FOMC meeting was held in 1936 firstly set by the Banking Act and the procedure of the voting was approximately tha same as today («for» and «against» the directive interest rate). At that time the document, which was available to the public was called the «Records of Policy Actions». It was published once a year and included only a few chapters explaining the reasoning behid the FOMC decisions.

The first move to the monetary policy transparency happened in June 1967 when it was decided to publish the «Records of Policy Actions» 90 days after the FOMC meeting. In the same year it was also announced about the public availability of the document the «Minutes of Actions», which contained the main discussions and votes during the procedure. There was also the document named the «Memorandum of Discussion», which contained private and detailed information about the meeting and which was not published. In 1975 the Committee announced that it reduces the time of the «Records of Policy Actions» from 90 to 45 days. Moreover, the next year the larger version of the «Records of Policy Actions» began to be published after the subsequent meeting.

In the early 1990s the new public document instead of «Records of Policy Actions» and «Minutes of Actions» was decided to be published. It was called «Minutes of the FOMC Meeting». Many researchers see it as the main step in the way of developing the monetary policy: «...transcript publication is only one element of procedural transparency and it is not by any means the most important element on the list…The Fed's decision in 1993 to begin publishing transcripts of FOMC meetings turned out to be the first in a series of transparency modifications that, in the aggregate, have significantly improved its communication with the public».

During the 1990's there was a slight and gradual tendency of making FOMC meetings more frequent, regular and stable. In 1994 there was an announcement about the publication of detailed policy decision explanations and changes in target market interest rate with the time lag. In 1995 the decisions about the immediate publication about only the final decision was made. In 1999 the publishment of bias in the directives accepted during the meeting was announced. In 2000 year the Committee began to announce the predictions about future aims and goals and economic development in a document about the balance of risks.

In 2002 the FOMC announced the publishment of the voters' names and preferred policy of any participants of the meeting including voting and non-voting members. In 2005 the minutes are supposed to be published after 3 weeks from the FOMC meeting.

The short and obvious scheme of the summaries of meetings is perfectly given by Deborah J. Danker and Matthew M. Luecke in their paper the «Background on FOMC Meeting Minutes»:

Pic. 1

History of dissents

Many reserchers suppose that the dissent of the Federal Open Market Committee has on average the least outstanding value of dissent among the other analyzed countries - very often the voting individuals stick to the Chairman's opinion instead of reflecting their real opinions: «The Fed's FOMC does vote in a formal sense, but it is widely known that individual members often do not vote their true preference. Instead, each committee member decides whether to support or oppose the chairman's policy recommendation, which is almost always made first. And Fed traditions dictate that a member should “dissent” only if he or she finds the majority's (that is, the chairman's) opinion unacceptable».

This statement may be approved by the statistical evidence from the appearence of FOMC in 1936 to nowadays. During the first twenty years there was nearly no dissent among the voting members relying on the available results. However, after the 1956 there was announced a more frequent tendency of FOMC meetings. The meetings began to be held more and more often, giving more opportunities for the dissents to arise. Relying on the graph shown in the Appendix 2 it can be concluded that analyzing the period from 1957 when the dissents began to appear to 2013, the largest number of dissents per year was in the periods from 1962 to 1965 and from 1978 to 1980.

During the last several decades more and more researches and papers are dedicted to the outlying evidence of the improvement of ability to predict market interest rates because of the publicly available information about minutes, extent of disagreement among the committee members and statements. However, the officially known dissent is said to be not the perfect indicator of the forecastibility of the market interest rate: «According to longstanding FOMC tradition, for example, a member is expected to vote in favor of the chairman's policy proposal unless he or she disagrees with it fundamentally--which is a much sterner test than merely preferring an alternative...More generally, the number of dissenting votes clearly underestimates the amount of disagreement».

That is why in my statistical research I will try to estimate the significance of not the only one, but different measures of dissents during the FOMC meeting in order to prove their benefitial power for the predictions of actual interest rates.



Chapter 3. The monetary policy in the United States of America

3.1 Aims

The monetary policy significance is becoming more and more obvious due to growing number of instabilities and crisises in the world financial markets. Most contemporary researchers consider the monetary policy decisions and their transparency to be vital step in overcoming a great amount of the difficulties economies face nowadays. The Federal Reserve System is responsible for the conducting of monetary policy in the United States of America, while the Federal Open Market Committee (which is one of the Fed parts) plays one of the most important roles in solving problems connected with the policy decisions.

The policymakers usually define several objectives for the monetary policy conduction in the USA and in other countries (also listed by Frederic Mishkin 2004 in «The economics of Money, Banking, and Financial markets»):

Stable interest rates of the economy. If there exists any often fluctuations of interest rates may lead to further instabilities and difficulties

Economic growth provision. This result may be achieved by increasing incentives for the suppliers to produce and for buyers to consume

Increase in the employment rates. The employment improvement can lead to the increasing output levels. The main aim of the government should be to stay at the natural rate of unemployment

Providing the financial markets stability. The interest rates changes may affect the stability of financial markets, the instability of which may lead to crisises and bankruptcies

Providing the foreign exchange markets stability. Such form of stability allows the domestic suppliers to become more competitive for the foreign markets and they have more oppurtunities to make their strategies in advance

Choosing the best combination of objectives to reach the best monetary policy direction

3.2 Monetary policy and the economy

Open market operations serve as one of the main instruments of the monetary policy discussed in this research. Such operations are conducted through the influence on the federal funds rate.

The Federal Reserve Banks, which are included into the Federal Reserve System, have their own balance sheets as common commercial banks. Changes in every constituent of assets or liabilities of Fed may influence the money supply.

The Fed assets:

Discount loans

Securities

Gold and SDR certificate accounts

Coin

Cash in collection

Other assets

The Fed Liabilities:

Currency outstanding

Foreign and other deposits

The USA Treasury deposits

Reserves

Deferred-availability cash items

Other capital accounts

Lynn S. Fox, Chair, Scott G. Alvarez in their publishment «The Federal Reserve System. Purposes and Functions» properly describe the monetary policy process in the USA:

The changes in federal funds rate can affect the list of different macroeconomic indicators like exchange rates of dollar, stock prices and can determine all other market interest rates. As a result, such variables as consumer and business spending are affected, shifting the Aggregate Demand curve outwards. Market participants are advised to follow data statements about the FOMC meeting decisions. It is stated that short-term interest rates are influenced positively by the news about federal funds rate changes. In their turn long-term rates depend on the short-term rates and their expectations. Such interest rates have their influence on the desire of investors to purchase stocks affecting the equity prices. Higher stock prices may influence positively the consumers' wealth. What's more, there is also an influence on the exchange rate of a dollar (when interest rates rise then there is a capital inflow, leading to an increase of dollar exchange rate).

Depending on the economy state, the monetary policymakers can obtain the following types of the policy:

Ease monetary policy. Such policy is also called the stimulating one and it is obtained in order to escape from the unemployment and low levels of output.

Tightening monetary policy. During the «overheated» state of the economy such policy helps to get rid of inflation.

However, there exists several problems like changes uncertainties, problems of lags and measurement bias, which may decrease the effectiveness of such policies.

3.3 Interest Rates' Dynamics

As it was said before, the Federal Funds Target Rate is the main determinant of the other market interest rates in the United States of America. Moreover, it plays one of the most significant roles for the economy as a whole. This rate is determined by the FOMC meeting, which is held several times a year since 1936, with the help of voting. As discussed previously, the targeting of the federal funds rate is one of the most commonly used in the USA ways of conducting necessary monetary policy. The federal funds rate is defined as the interest rate set by banks for the overnight loans they charge each other for the borrowing of Fed funds.

However, the Federal Funds Target Rate is not always equal to the real market Effective Federal Funds Rate, which is calculated as the average rate for all the actions including the borrowing of funds between the banks.

The prime loan rate is determined as a rate at which commercial banks make short-term loans to the most relient borrowers. It is not set by the FOMC meetings, but this rate is set depending on the target level of the federal funds rate. The prime rate directly influences such variables as interest mortgages, credit card rates, loan rates for individuals and small businesses. In the USA for the last several decades the prime loan rate is 3% more than the Federal Funds Target Rate.

Looking at the graph (Appendix 3) it becomes possible to analyze the dynamics of the described interest and prove their dependence on the Federal Funds Target Rate. It can be seen that for the period from 1936 to 2006 the Effective Federal Funds Rate almost always coincides with the Federal Funds Target Rate. While the prime loan rate, having the same dynamics as the Federal Funds Target Rate, is constanly around 3% higher. That is why it is possible to say that the prime loan rate directly depends on the Federal Funds Target Rate. Thus, the federal funds target rate determines the value of the prime loan rate, that is why, it is possible to use the dissents from the FOMC meetings for the testing the prime loan interest rate predictability.



Chapter 4. Statistical analysis of the dissent

4.1 Data collected

The whole analysis will be conducted using the data from September 1987 to January 2002. Such time period my be explained by the fact that only in January 2002 the Federal Reserve System began to public detailed reports on the voting patterns (including names of voters, their opinions during the meeting and the statement) and the aim of this research is to determine whether the knowledge of these statements would have improved the forecasting power of market interest rate before the 2002.

Thus, in this research the hypothesis that if such information about the meeting was published before 2002, the forecasters would have been better off is tested. Using the data on the FOMC meetings and the statements published by the Federal Reserve System on the dissent for the given period, this paper analyzes whether the individual policy preferences revealed during policy deliberations at the FOMC meetings bring benefit to the regression.

The significance of the dissents researched further in this paper arises because of the several reasons. Firstly, different measures of dissent take into the account the opinions of the participants of the meeting that are unable to vote. Such factor as their policy can be a great forecast of the future interest rate. Moreover, there exists the possibility that voters initially wanted to stick to one opinion, but in the process of the discussion they changed their decisions. Such change is also taken into the account by the unofficial dissent and may improve the fit of the regression and the interest rate predictability. What is more, such types of the dissents are thought to take into account the real voters' preferences. For example, very often an individual don't want to have a voting result different from the Chairman or from his/her colleagues, while the dissents researched in this paper take into the account the minimal fluctuations of the opinions.

Thus, for every FOMC meeting the unofficial rate of dissent was counted. The information about personal preferences of each participant of the meeting was derived from the official reports and statements published by the Federal Reserve System.

If we consider a committee with N members, for each i member and each policy setting meeting t let's define the given indicator function:

Where is the change to the policy rate prefered by member i and is the change made by the committee. This definition includes each voter's preference over the suggested alternative. If the member wanеed to make the change in the interest rate less (more) than the suggested by the committee, than the indicator function is equal to -1 (1). In the case of the coincidence of member's desire and the committee decision the indicator function is to be equal to zero. The given function doesn't take into consideration the relationship between a concrete voter (his name, status and district) and his voting results.

Thus, the final measure of dissent is represented as the sum of indicator functions for each member of the commitee divided by the number of members.

Based on the available data the following individual policy preferences revealed during policy deliberations at the FOMC meetings can be mentioned:

Dissent among all members, including voting and non-voting participants (DissA)

Dissent among voting members only (DissV)

Dissent among those who will be voting at next meeting (DissV2)

Thus, for every FOMC meeting for the time period 1987-2002 data about different measures of dissent was collected. Overall 453 different results of different kinds of dissent were collected (three dissents for every meeting). During this period 133 FOMC meetings were held (almost every month). The source of this data is the research of Andrei Sirchenko.

On the diagram given below we can analyze how the given dissents vary in absolute values over the given period (Appendix 4). It can be concluded that the Dissent among all members (DissA) is the most fluctuating variable and, thus, it may bring the largest benefit to the regression.

This research also requires data about actual and realized US prime loan rates (later in methodology as Ya). The source of such data is the Federal Reserve Bank of St. Louis. Only average quarterly rates were taken, as the forecasts published are made for the current and the following quarters.

The Blue Chip Financial Forecasts serve as a source of the 141 individual forecasts of the prime loan rate of the USA. They represent the forecasts of famous analytical agencies, which publish their forecasts every month for four quarters (including the current) as an average rate for the quarter. The forecasts are published on the first day of every month.

In order to transform all the data into the form in which it would be possible to test the significance of the individual policy preferances for the regression, all the results are presented as monthly parameters. Actual US prime loan rates are presented as monthly by making the average quarterly rates monthly by making the rate equal every three months (taking the average quarterly for every month of this quarter).

Moreover, the fact that FOMC meeting results could have been considered at the next day and immediately used for the forecasting is taken into consideration. Thus, in this paper it is assumed that if the meeting was held before the 20th day of the month (say, on the 12th of April), then forecasters have an opportunity to work out the information for their forecasts that would be published on the 1st of May. However, if the FOMC meeting results were published, for example, on the 26th of June, then the individual forecasters would only have an opportunity and time to properly use the information from the previous meeting (May) for their July forecasts.

Thus it can be concluded that using the given data set of macroeconomic indicators and forecasts it is now possible to continue with an appropriate analysis.

4.2 Methodology

The whole practical analysis of the paper would be divided into two parts: using aggregated data and disaggregated data (panel data). Both models would be used in order to test whether the different types of individual policy preferences revealed during policy deliberations at the FOMC meetings are significant for the regression.

Aggregate data

According to Blue Chip Financial Forecasts forecasters make their predictions for the next several quarters including the current one. In this research the cases when k=0,1 and 2 would be analyzed. This means that for every k (three regressions) the regressions on all valid kinds of dissent would be analyzed:

K=0 - when the forecast was made for the current quarter

K=1 - when the forecast was made for the next quarter

K=2 - when the forecast was made for the quarter after the next one

Variable c represents the constant term of the regression.

is the variable which is considered to be dependent. It represents the historical actual and realized monthly US prime loan rates for the period from 1987 to 2002. The index k means that if k=1 then, for example, if we take March as the date of making the forecast (t), we should take the actual value of interest rate of the next quarter (t+1).

is the variable which is represented as the average (among individual forecasters) forecasted value of the interest rate made in period t for the period t+k. For example, if k=2 and the forecast is made in January (t) then this value represents the forecast made in January for the second quarter (t+2).

is the variable which represents all the kinds and types of diverse dissents described in the previous section. The regression would be tested towards each of that dissents separately including each ot them in order to find out whether the given measures of dissent are significant for the dissent regression and which type is the most appropriate one for the improving the forecasting power of the market interest rate.

Disaggregated data

The main focus of this research would be made on the panel data analysis. It would cover a great variety of different techniques in order to analyze panel data. «Panel data (also known as longitudinal or cross-sectional time-series data) is a dataset in which the behavior of entities are observed across time. These entities could be states, companies, individuals, countries, etc». The panel of my research is considered as unbalanced, because there doesn't exist all the observations for every unit of observation of every time period. The main advantages of panel data are:

One of the main advantages of panel data over the aggregated data is that it gives an opportunity to measure the variables changing over time but being constant over entities.

Panel data gives a researcher a possibility to consider more observations. Thus, it helps to increase the degrees of freedom, reduce the correlation between the explanatory variables and decrease the standard errors.

It also takes into account different factors, which are impossible to calculate (sex, culture).

Panel data helps to overcome the problem of impossibility to analyze the regressions using the multilevel modeling.

Such data also gives an opportunity to avoid the specification errors like not including important variables into the regression

Panel data is usually said to be of higher quality due to well designed surveys and data collections

Unlike the cross-section data panel data allows to analyze dynamics of observations more explicitly

However, panel data also have some disadvantages. They include:

Non random data sample data (missing data for the dependent variable) which leads to the self-selection bias

Heterogeneous bias, which results from the ignorance of heterogeneous parameters and leads to the inconsistency of estimates

Correlation between countries in macro panels (in this research between forecasters)

The model used in this research has the following form:

As in the previous paragraph is the variable which is considered to be dependent. It represents the historical actual and realized monthly US prime loan rates for the given period (from 1987 to 2002). The index k means that if the forecast was made in zero period for the k period (quarter), the actual value of interest rate would be drawn from the kth period and would be for every forecaster.

The variable represents the constant term of each entity of forecasters(from 141).

is the variable which is represented as the forecasted value of the interest rate made in period t for the period t+k for every forecaster entity (unlike the aggregated data where it is represented as the average among individual forecasters).

as described in the previous section is the variable which represents different types of the dissents. The panel regression would be also tested towards each type of the dissent and tested for their significance.

The main techniques I will use in my further analysis are fixed and random effects.

Fixed effects

Such model technique is said to be the best one if during the analysis the researcher wants to focus on the unique sample of N firms (in our case on 141 forecasters) and the conclusions would be drawn depending on their behavior. We consider how the variables vary over the given period of time within a certain entity each with its own features and patterns. Fixed effect model gives an opportunity to transform the model in a way to reduce the unobserved heterogeneity term, which consists of unobserved variables not varying over time. Such technique assumes that every entity analyzed is considered to be different with its own time-invariant characteristics uncorrelated with the other individual ones. If the error term of one entity is related to the error term of the another entity then Fixed Effect technique is not suitable.

«The fixed-effects model controls for all time-invariant differences between the individuals, so the estimated coefficients of the fixed-effects models cannot be biased because of omitted time-invariant characteristics...(like culture, religion, gender, race, etc)».

For the estimation of panel data using fixed effects a lot of ways may be used. The examples are Least squares dummy variables (LSDV) method and the Within groups method.

The LSDV method is based on the introduction of dummies into the panel regression. Using this research as an example, dummy variables should be added for each forecaster, thus, the number of dummies will be equal to 141. Such dummies allow the researcher to estimate the pure effect of variables in panel. The panel regression would be of the following form:



Where all the variables are decribed as in the previous equation, while

The within groups method allows to escape the unobserved heterogeneous term. «This method is called «within groups» one because the variations of the dependent variable around its mean are regressed on the variations of explanatory variables around their means». For the large amount of observations these two methods are thought as similar to each other.

Random effects

Random effect is usually thought to be more consistent than the Fixed Effect. Unlike the Fixed one, the Random takes into consideration the variability of the variables not over the whole sample, but within each certain entity. Moreover, such effect includes the variables that are constant in time and they serve as explanatory variables.

A Hausman test usually serves a way to determine the effect, which can be used towards the given regression. The random effect can be used only in the case of non-correlation of random effects with the variables, thus, it happens not often. The zero hypothesis states that errors are uncorrelated with regressors (the Random effect), while the alternative states that such relationship exists (the Fixed Effect).

In the further research I would also use a Wald test, which compares the Fixed Effect with the pooled panel. It allows to determine whether there exists any individual effects. Thus, the null hypothesis states that the panel is pooled, while the alternative hypothesis is that the effect is fixed.

The statistical analysis of the influence of dissent on the predictability of market interest rates

For the statistical testing I selected the following types of dissents because of the most accurate and detailed information collected: dissent among all members, dissent among voting members only and dissent among those who will be voting at next meeting. Thus, for every k (0,1,2) there would be tested three regressions analyzing the significance of the dissent coefficients and showing, which of them would improve the fit of the regression. All of the time series used in further research were checked for stationarity with the usage of ADF (augmented Dickey Fuller) unit root tests. For all data used the null hypothesis of no stationarity was rejected at 5% significance level. The results can be observed in Appendix 5.

Aggregate data testing

Having tried variety of different regressors, the final form of the regression analyzed would be: , where the variable would be DissA, DissV or DissV2 (definitions given in the previous paragraph).

Continuing with the further analysis, all the regressions for k=0, 1 and 2 were respectfully analyzed. The results can be seen in the Appendix 6. After analyzing the results, it can be concluded that almost in all the regressions the significance of all types of dissent coefficients can be proved by using t-statistics at a 5% significance level (the null hypothesis of the each dissent coefficient insignificance is rejected). The next step of the analysis was to provide the Breusch-Godfrey test on serial correlation for each pair of regressions for k=0, 1 and 2 (overall nine tests). The zero hypothesis of this test states that there is no serial correlation, while the alternative one states that the correlation exists. The results of all tests showed that the zero hypothesis is rejected (Appendix 7). Thus, I used a Newey-West estimator in order to get rid of autocorrelation and heteroskedasticity in the error terms. The main results of t-statistics for the DissA, DissV and DissV2 coefficients may be summarized in the following table:

	K=0	K=1	K=2
	t-statistic	p-value	t-statistic	p-value	t-statistic	p-value
DissA	1,808589	0,0723	2,382034	0,0183	2,580902	0,0107
DissV	1,510157	0,1329	1,884755	0,0612	1,802053	0,0733
DissV2	1,562776	0,1200	1,862055	0,0643	1,755382	0,0810



The more detailed results can be seen using the Appendix 8.

The main conclusions that can be driven from these results say that after applying a Newey-West estimator variables DissV and DissV2 loose their significances for the regression. If we analyze the DissA measure, then for the forecasts, which were made for the next or for the one after the next quarters, its value would be significant for the regression (at 5% significance level) and, thus, would improve the fit of the regression and the predictability of the market interest rate. The insignificance of all measures of dissent for the forecasts made for the current quarter may be explained by the fact that the market doesn't fully adjust to the information released in the current quarter and may not be able to stick to the predictions made.

Disaggregate data testing

The regression used in the panel data testing will be as follows:

Providing an analysis of all the regressions using the Fixed and Random effects and looking at the dissent coefficients significance I decided to continue the analysis of the panel data with the same types of dissents as in the case of the aggregated data: dissent among all members, dissent among voting members only and dissent among those who will be voting at next meeting.

The first test conducted is the Hausman test, which allows to choose the Fixed or Random effect to be used in the model. This test was conducted for each of the nine regressions. If the p-value < 0,01 then the zero hypothesis is rejected. Summarizing the test results:

	K=0	K=1	K=2
	P-value	P-value	P-value
DissA ...

Подобные документы

• The create a system of predictive mathematical analysis

  The air transport system in Russia. Project on the development of regional air traffic. Data collection. Creation of the database. Designing a data warehouse. Mathematical Model description. Data analysis and forecasting. Applying mathematical tools.
  
  реферат [316,2 K], добавлен 20.03.2016
  

• The stock market of Russia and Italy: dynamics of development and comparative characteristics

  The stock market and economic growth: theoretical and analytical questions. Analysis of the mechanism of the financial market on the efficient allocation of resources in the economy and to define the specific role of stock market prices in the process.
  
  дипломная работа [5,3 M], добавлен 07.07.2013
  

• Economic Policy of the United Kingdom in the crisis of 2008-2010

  The global financial and economic crisis. Monetary and financial policy, undertaken UK during a crisis. Combination of aggressive expansionist monetary policy and decretive financial stimulus. Bank repeated capitalization. Support of domestic consumption.
  
  реферат [108,9 K], добавлен 29.06.2011
  

• Stabilization policy: macroeconomic dimensions

  Models and concepts of stabilization policy aimed at reducing the severity of economic fluctuations in the short run. Phases of the business cycle. The main function of the stabilization policy. Deviation in the system of long-term market equilibrium.
  
  статья [883,7 K], добавлен 19.09.2017
  

• Electronic Construction Procurement System

  Project background and rationales. Development methodology, schedule planning. Company mission and vision. Organization of staff and company structure. Procurement system target market. Implementation of procurement system. Testing, user manual.
  
  дипломная работа [6,8 M], добавлен 28.11.2013
  

• Economy of the United States of America

  Socio-economic and geographical description of the United states of America. Analysis of volumes of export and import of the USA. Development and state of agroindustrial complex, industry and sphere of services as basic sectors of economy of the USA.
  
  курсовая работа [264,5 K], добавлен 06.06.2014
  

• Market and its Functioning

  Law of demand and law of Supply. Elasticity of supply and demand. Models of market and its impact on productivity. Kinds of market competition, methods of regulation of market. Indirect method of market regulation, tax, the governmental price control.
  
  реферат [8,7 K], добавлен 25.11.2009
  

• The market analysis of natural gas in the European Union

  Natural gas market overview: volume, value, segmentation. Supply and demand Factors of natural gas. Internal rivalry & competitors' overview. Outlook of the EU's energy demand from 2007 to 2030. Drivers of supplier power in the EU natural gas market.
  
  курсовая работа [2,0 M], добавлен 10.11.2013
  

• Formation of the labor market in Ukraine

  Economic entity, the conditions of formation and functioning of the labor market as a system of social relations, the hiring and use of workers in the field of social production. Study of employment and unemployment in the labor market in Ukraine.
  
  реферат [20,3 K], добавлен 09.05.2011
  

• Economic Analysis of Italy and United States

  Short and long run macroeconomic model. Saving and Investment in Italy, small open economy. Government expenditure and saving scatterplot. Loanable market equilibrium in closed economy in the USA. Okun’s Law in the USA and Italy, keynesian cross.
  
  курсовая работа [1,6 M], добавлен 20.11.2013
  

• Private sector and human-resource development in Georgia

  Government’s export promotion policy. Georgian export promotion agency. Foreign investment promotion. Government’s foreign investment promotion policy. Foreign investment advisory council. Taxation system and tax rates in Georgia.
  
  курсовая работа [644,0 K], добавлен 24.08.2005
  

• The significance of supply and demand for the market

  A theoretic analysis of market’s main rules. Simple Supply and Demand curves. Demand curve shifts, supply curve shifts. The problem of the ratio between supply and demand. Subsidy as a way to solve it. Effects of being away from the Equilibrium Point.
  
  курсовая работа [56,3 K], добавлен 31.07.2013
  

• Chinese Labor Market

  Transition of the Chinese labor market. Breaking the Iron Rice Bowl. Consequences for a Labor Force in transition. Labor market reform. Post-Wage Grid Wage determination, government control. Marketization Process. Evaluating China’s industrial relations.
  
  курсовая работа [567,5 K], добавлен 24.12.2012
  

• The main principles of safety market economy

  Assessment of the rate of unemployment in capitalist (the USA, Germany, England, France, Japan) and backward countries (Russia, Turkey, Pakistan, Afghanistan). Influence of corruption, merges of business and bureaucracy on progress of market economy.
  
  реферат [15,5 K], добавлен 12.04.2012
  

• Effectiveness of merger and acquisition deals in Russia

  Mergers and acquisitions: definitions, history and types of the deals. Previous studies of post-merger performance and announcement returns and Russian M&A market. Analysis of factors driving abnormal announcement returns and the effect of 2014 events.
  
  дипломная работа [7,0 M], добавлен 02.11.2015
  

• The main results of development of Kazakhstan in the years of independence

  The first stage of market reforms in Kazakhstan is from 1992 to 1997. The second phase is in 1998 after the adoption of the Strategy "Kazakhstan-2030". The agricultural, education sectors. The material and technical foundation of the medical institutions.
  
  презентация [455,3 K], добавлен 15.05.2012
  

• Assessment of the situation on the regional housing market in Russia

  Issues about housing prices formation process. Analytical model of housing prices. Definition a type of relationship between the set of independent variables and housing prices. The graph of real housing prices of all Russian regions during the period.
  
  курсовая работа [1,6 M], добавлен 23.09.2016
  

• Financial bubble

  Financial bubble - a phenomenon on the financial market, when the assessments of people exceed the fair price. The description of key figures of financial bubble. Methods of predicting the emergence of financial bubbles, their use in different situations.
  
  реферат [90,0 K], добавлен 14.02.2016
  

• Application requirements of international institutions to organizations of financial monitoring /AML in Ukraine in the struggle against corruption

  The influence of corruption on Ukrainian economy. Negative effects of corruption. The common trends and consequences of increasing corruption. Crimes of organized groups and criminal organizations. Statistical data of crime in some regions of Ukraine.
  
  статья [26,7 K], добавлен 04.01.2014
  

• Economic efficiency of grain production and its influence on a condition of the agricultural enterprise

  The essence of economic efficiency and its features determination in grain farming. Methodology basis of analysis and efficiency of grain. Production resources management and use. Dynamics of grain production. The financial condition of the enterprise.
  
  курсовая работа [70,0 K], добавлен 02.07.2011
  

• главная
• рубрики
• по алфавиту
• вернуться в начало страницы
• вернуться к началу текста
• вернуться к подобным работам