Оценка факторов, влияющих на рыночную стоимость интернет-компаний

4 817

2004

Россия

17

YY

YY Inc

3 092

2005

Китай

· Сегмент «Провайдеры контента»

#	Тикер	Компания	Капитализация, млн USD, 2015	Год основания	Страна
1	ANGI	Angie's List Inc	343	1995	США
2	AOL	AOL Inc	3 106	1993	США
3	CSGP	CoStar Group Inc	6 392	1987	США
4	DHX	Dice Holdings Inc	488	1996	США
5	DMD	DEMAND MEDIA	113	2006	США
6	GOGO	Gogo Inc	1 626	1991	США
7	MWW	Monster Worldwide Inc	576	1994	США
8	NFLX	Netflix Inc	25 209	1997	США
9	O1BC	XING AG	898	2003	Германия
10	P	Pandora Media Inc	3 390	2000	США
11	RATE	Bankrate Inc	1 184	1976	США
12	SINA	SINA Corp/China	2 143	1999	Китай
13	SSTK	Shutterstock Inc	2 449	2003	США
14	STMP	Stamps.com Inc	1 098	1996	США
15	WBMD	WebMD Health Corp	1 631	1995	США
16	YOKU	YOUKU.COM INC	2 475	2006	Китай

· Сегмент «Электронная коммерция»

#	Тикер	Компания	Капитализация, млн USD, 2015	Год основания	Страна
1	8044 TT	PCHOME ONLINE INC	35 334	1996	Китай
2	AMTD	TD Ameritrade Holding Corp	20 252	1971	США
3	AMZN	Amazon.com Inc	172 797	1994	США
4	AWAY	HomeAway Inc	2 852	2004	США
5	CASH	CASH (Meta Financial Group)	255	1954	США
6	CTRP	Ctrip.com International Ltd	7 932	1999	Китай
7	EBAY	eBay Inc	69 946	1995	США
8	ETFC	E*TRADE Financial Corp	8 277	1982	США
9	EXPE	Expedia Inc	11 928	1996	США
10	GRPN	Groupon Inc	4 869	2008	США
11	LQDT	Liquidity Services Inc	296	1999	США
12	MELI	MercadoLibre Inc	5 410	1999	Аргентина
13	NILE	Blue Nile Inc	378	1999	США
14	NTRI	Nutrisystem Inc	579	1972	США
15	OSTK	Overstock.com Inc	588	1997	США
16	OWW	Orbitz Worldwide Inc	1 298	2000	США
17	PCLN	Priceline Group Inc/The	60 465	1997	США
18	SFLY	Shutterfly Inc	1 708	1999	США
19	TZOO	Travelzoo Inc	142	1998	США
20	XOOM	Xoom Corp	568	2001	США
21	Z	Zillow Group Inc	5 779	2004	США

· Сегмент «Программное обеспечение»:

#	Тикер	Компания	Капитализация, млн USD, 2015	Год основания	Страна
1	2432 JT	DENA CO LTD	354 554	1999	Япония
2	3632 JT	GREE INC	200 951	2004	Япония
3	3659 JT	Nexon Co Ltd	552 265	1994	Япония
4	AKAM	Akamai Technologies Inc	12 705	1998	США
5	BCOR	Blucora Inc	561	1996	США
6	BCOV	Brightcove Inc	238	2004	США
7	BV	Bazaarvoice Inc	449	2005	США
8	CARB	Carbonite Inc	390	2005	США
9	CCIH	ChinaCache International Holdings Ltd	301	1998	Китай
10	CCOI	Cogent Communications Holdings Inc	1 639	1999	США
11	CRM	salesforce.com inc	43 467	1999	США
12	CSOD	Cornerstone OnDemand Inc	1 557	1999	США
13	CTCT	Constant Contact Inc	1 228	1998	США
14	CYOU	CHANGYOU.COM LTD	1 368	2007	Китай
15	DWRE	Demandware Inc	2 223	2004	США
16	EBIX	Ebix Inc	1 071	1976	США
17	EGAN	eGain Corp	88	1997	США
18	EGOV	NIC Inc	1 169	1991	США
19	ELNK	EarthLink Holdings Corp	457	1994	США
20	EQIX	Equinix Inc	13 150	1998	США
21	INAP	Internap Corp	557	1996	США
22	JCOM	j2 Global Inc	3 168	1995	США
23	JIVE	JIVE SOFTWARE INC	382	2001	США
24	JNPR	Juniper Networks Inc	9 190	1996	США
25	LOGM	LogMeIn Inc	1 370	2003	США
26	LPSN	LivePerson Inc	583	1995	США
27	MCHX	Marchex Inc	171	2003	США
28	MDRX	Allscripts Healthcare Solutions Inc	2 159	1986	США
29	MRIN	Marin Software Inc	231	2006	США
30	N	NetSuite Inc	7 171	1998	США
31	NQ	NQ Mobile Inc	231	2005	Китай
32	NTGR	NETGEAR Inc	1 141	1996	США
33	OTEX	Open Text Corp	6 462	1994	США
34	PERI	Perion Network Ltd	237	1999	Израиль
35	PRFT	Perficient Inc	729	1997	США
36	QIHU	Qihoo 360 Technology Co Ltd	6 447	2005	Китай
37	QNST	QuinStreet Inc	265	1999	США
38	RAX	Rackspace Hosting Inc	7 299	1998	США
39	SCOR	comScore Inc	1 747	1999	США
40	SONS	Sonus Networks Inc	390	1997	США
41	SQI	SciQuest Inc	468	1995	США
42	TRAK	Dealertrack Technologies Inc	2 094	2001	США
43	VNET	21Vianet Group Inc	1 465	1999	Китай
44	VRSN	VeriSign Inc	7 828	1995	США
45	WIFI	Boingo Wireless Inc	274	2001	США
46	WWWW	Web.com Group Inc	982	1999	США
47	ZIXI	Zix Corp	224	1988	США
48	ZNGA	ZYNGA INC	2 598	2007	США

Приложение 2

Список переменных статистической модели

Зависимая переменная:

PX_LAST

· PX_LAST - цена акции при закрытии, в долл. США

Независимые переменные:

· IS_EPS - Чистая прибыль на обыкновенную акцию, в долл. США

· REVENUE_PER_SH - Выручка на обыкновенную акцию, в млн. долл. США

· EBITDA_PER_SH - EBITDA (прибыль до вычета расходов по выплате процентов, налогов и начисленной амортизации) на обыкновенную акцию, в млн. долл. США

· FREE_CASH_FLOW_PER_SH - Чистый денежный поток на обыкновенную акцию, в млн. долл. США

· BOOK_VAL_PER_SH - Балансовая стоимость компании на обыкновенную акцию, в млн. долл. США

· VISITORS_PER_SH - количество посетителей сайта за период, нормированное по количеству обыкновенных акций в обращении, в ед.

· BOUNCE_RATE - Показатель отказов: процентное соотношение количества посетителей, покинувших сайт прямо со страницы входа или просмотревших не более одной страницы сайта, в процентах

· DURATION - Среднее время, проведенное на сайте одним посетителем, в сек.

· PAGEVIEWS - Количество просмотренных страниц на одного посетителя сайта, ед.

Приложение 3

Результаты регрессионного анализа по сегменту «социальные сети и поисковые системы»

Модель 1. y ~ f1 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-90.856 -24.638 -2.436 35.737 78.095

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.628e+03 7.990e+02 2.037 0.0566 .

f1 2.794e+01 1.601e+01 1.745 0.0981 .

q1 -2.917e+01 5.869e+01 -0.497 0.6252

q2 -9.503e-01 1.002e+00 -0.948 0.3557

q3 -3.793e+02 5.620e+02 -0.675 0.5083

q4 -2.876e-05 5.710e-06 -5.036 8.6e-05 ***

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 54.15 on 18 degrees of freedom

Multiple R-squared: 0.8874, Adjusted R-squared: 0.8561

F-statistic: 28.36 on 5 and 18 DF, p-value: 6.135e-08

Модель 2. y ~ f2 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-87.716 -33.618 0.907 36.966 90.258

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 8.551e+02 8.529e+02 1.003 0.3294

f2 -2.679e+00 6.315e+00 -0.424 0.6765

q1 7.514e+00 6.267e+01 0.120 0.9059

q2 -1.941e-01 9.859e-01 -0.197 0.8461

q3 -4.465e+02 6.229e+02 -0.717 0.4827

q4 -2.221e-05 6.867e-06 -3.235 0.0046 **

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 58.26 on 18 degrees of freedom

Multiple R-squared: 0.8696, Adjusted R-squared: 0.8334

F-statistic: 24.01 on 5 and 18 DF, p-value: 2.228e-07

Модель 3. y ~ f3 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-89.794 -37.016 0.362 36.299 86.740

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.129e+03 8.266e+02 1.365 0.188965

f3 -1.064e+01 1.189e+01 -0.894 0.382975

q1 -6.048e-01 5.913e+01 -0.010 0.991951

q2 -5.438e-01 1.064e+00 -0.511 0.615614

q3 -4.348e+02 5.974e+02 -0.728 0.476121

q4 -2.434e-05 5.879e-06 -4.141 0.000614 ***

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 57.3 on 18 degrees of freedom

Multiple R-squared: 0.8739, Adjusted R-squared: 0.8389

F-statistic: 24.95 on 5 and 18 DF, p-value: 1.659e-07

Модель 4. y ~ f5 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-87.861 -36.393 0.598 36.501 84.869

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.184e+03 8.090e+02 1.464 0.160

f5 -1.071e+01 1.048e+01 -1.022 0.321

q1 -8.427e+00 6.084e+01 -0.139 0.891

q2 -5.653e-01 1.035e+00 -0.546 0.592

q3 -4.805e+02 5.985e+02 -0.803 0.432

q4 -2.350e-05 4.511e-06 -5.210 5.9e-05 ***

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 56.93 on 18 degrees of freedom

Multiple R-squared: 0.8755, Adjusted R-squared: 0.8409

F-statistic: 25.32 on 5 and 18 DF, p-value: 1.481e-07

Модель 5. y ~ f6 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-86.491 -34.210 4.051 36.801 91.982

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 7.619e+02 9.673e+02 0.788 0.44114

f6 -1.322e+00 5.871e+00 -0.225 0.82435

q1 9.007e+00 7.381e+01 0.122 0.90423

q2 -1.261e-02 9.408e-01 -0.013 0.98945

q3 -4.527e+02 6.812e+02 -0.665 0.51473

q4 -2.096e-05 7.037e-06 -2.978 0.00806 **

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 58.47 on 18 degrees of freedom

Multiple R-squared: 0.8687, Adjusted R-squared: 0.8322

F-statistic: 23.81 on 5 and 18 DF, p-value: 2.373e-07

Приложение 4

Результаты регрессионного анализа по сегменту «провайдеры контента»

Модель 1. y ~ f1 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-14.2389 -4.5030 -0.2943 3.9433 14.2774

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 2.877e+02 6.708e+01 4.289 0.000442 ***

f1 2.196e+01 9.776e+00 2.247 0.037448 *

q1 -1.422e+01 7.913e+00 -1.797 0.089216 .

q2 -1.430e-01 1.703e-01 -0.840 0.412055

q3 -2.121e+02 7.562e+01 -2.804 0.011732 *

q4 -1.646e-05 7.954e-06 -2.070 0.053116 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.814 on 18 degrees of freedom

Multiple R-squared: 0.6148, Adjusted R-squared: 0.5078

F-statistic: 5.745 on 5 and 18 DF, p-value: 0.002436

Модель 2. y ~ f2 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-11.0910 -4.7140 0.0404 2.6742 20.3873

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 3.006e+02 8.304e+01 3.620 0.00196 **

f2 2.773e+00 5.520e+00 0.502 0.62153

q1 -1.888e+01 8.562e+00 -2.206 0.04064 *

q2 -1.875e-01 1.900e-01 -0.986 0.33696

q3 -2.403e+02 8.378e+01 -2.868 0.01024 *

q4 -2.344e-06 1.213e-05 -0.193 0.84893

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 8.78 on 18 degrees of freedom

Multiple R-squared: 0.5136, Adjusted R-squared: 0.3784

F-statistic: 3.801 on 5 and 18 DF, p-value: 0.01589

Модель 3. y ~ f3 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-10.846 -6.062 0.475 3.358 18.835

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 3.098e+02 7.211e+01 4.296 0.000435 ***

f3 9.095e+00 7.931e+00 1.147 0.266464

q1 -1.913e+01 8.313e+00 -2.302 0.033521 *

q2 -1.691e-01 1.854e-01 -0.912 0.373901

q3 -2.323e+02 8.176e+01 -2.841 0.010846 *

q4 -9.027e-06 7.587e-06 -1.190 0.249570

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 8.535 on 18 degrees of freedom

Multiple R-squared: 0.5403, Adjusted R-squared: 0.4126

F-statistic: 4.232 on 5 and 18 DF, p-value: 0.01016

Модель 4. y ~ f5 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-14.7443 -2.6026 0.5001 2.8168 10.7749

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 2.208e+02 6.396e+01 3.452 0.00284 **

f5 3.532e+01 1.013e+01 3.485 0.00264 **

q1 -1.153e+01 6.999e+00 -1.648 0.11676

q2 -5.883e-02 1.524e-01 -0.386 0.70400

q3 -1.971e+02 6.637e+01 -2.970 0.00821 **

q4 -8.518e-06 5.934e-06 -1.435 0.16833

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 6.832 on 18 degrees of freedom

Multiple R-squared: 0.7055, Adjusted R-squared: 0.6237

F-statistic: 8.623 on 5 and 18 DF, p-value: 0.0002594

Модель 5. y ~ f6 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-14.9188 -3.6659 -0.1434 2.8619 18.0501

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 2.508e+02 8.066e+01 3.110 0.00605 **

f6 1.121e+00 6.764e-01 1.657 0.11479

q1 -1.740e+01 8.086e+00 -2.151 0.04528 *

q2 -1.611e-01 1.789e-01 -0.900 0.37981

q3 -1.978e+02 8.270e+01 -2.392 0.02788 *

q4 4.161e-06 9.848e-06 0.422 0.67768

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 8.235 on 18 degrees of freedom

Multiple R-squared: 0.572, Adjusted R-squared: 0.4532

F-statistic: 4.812 on 5 and 18 DF, p-value: 0.005734

Приложение 5

Результаты регрессионного анализа по сегменту «электронная коммерция»

Модель 1. y ~ f1 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-13.5728 -3.1447 -0.0053 3.4768 15.8920

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.064e+02 8.691e+01 1.225 0.2364

f1 4.250e+01 6.848e+00 6.206 7.41e-06 ***

q1 -3.208e+01 5.856e+00 -5.478 3.34e-05 ***

q2 7.820e-01 1.470e-01 5.321 4.66e-05 ***

q3 -2.627e+02 1.312e+02 -2.002 0.0606 .

q4 2.588e-05 3.425e-05 0.756 0.4597

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.978 on 18 degrees of freedom

Multiple R-squared: 0.8383, Adjusted R-squared: 0.7933

F-statistic: 18.66 on 5 and 18 DF, p-value: 1.475e-06

Модель 2. y ~ f2 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-15.668 -6.205 1.692 3.360 14.076

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.685e+02 8.126e+01 2.074 0.05271 .

f2 1.193e+01 1.887e+00 6.320 5.89e-06 ***

q1 -2.297e+01 6.051e+00 -3.795 0.00133 **

q2 1.382e-01 1.425e-01 0.970 0.34470

q3 -2.152e+02 1.314e+02 -1.638 0.11873

q4 4.383e-05 3.480e-05 1.259 0.22394

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.88 on 18 degrees of freedom

Multiple R-squared: 0.8422, Adjusted R-squared: 0.7984

F-statistic: 19.22 on 5 and 18 DF, p-value: 1.187e-06

Модель 3. y ~ f3 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-11.5092 -2.3115 -0.0155 2.4092 13.4267

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 4.163e+01 8.194e+01 0.508 0.617568

f3 3.701e+01 5.127e+00 7.219 1.03e-06 ***

q1 -2.565e+01 5.387e+00 -4.762 0.000156 ***

q2 6.383e-01 1.253e-01 5.094 7.57e-05 ***

q3 -1.519e+02 1.217e+02 -1.249 0.227748

q4 5.950e-05 3.233e-05 1.840 0.082307 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.164 on 18 degrees of freedom

Multiple R-squared: 0.8696, Adjusted R-squared: 0.8334

F-statistic: 24.01 on 5 and 18 DF, p-value: 2.228e-07

Модель 4. y ~ f5 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-21.728 -5.451 -1.276 6.318 21.186

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 2.651e+02 1.186e+02 2.235 0.038308 *

f5 9.776e+00 3.524e+00 2.774 0.012505 *

q1 -3.485e+01 8.667e+00 -4.021 0.000802 ***

q2 5.211e-01 2.040e-01 2.554 0.019938 *

q3 -2.907e+02 1.973e+02 -1.474 0.157857

q4 3.068e-06 5.075e-05 0.060 0.952452

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 11.83 on 18 degrees of freedom

Multiple R-squared: 0.6443, Adjusted R-squared: 0.5455

F-statistic: 6.52 on 5 and 18 DF, p-value: 0.001262

Модель 5. y ~ f6 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-10.5000 -3.0933 -0.3044 1.9523 16.3722

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 1.235e+01 7.251e+01 0.170 0.8667

f6 8.414e+00 9.768e-01 8.613 8.43e-08 ***

q1 -7.481e+00 5.545e+00 -1.349 0.1941

q2 1.006e-02 1.175e-01 0.086 0.9327

q3 -5.136e+01 1.099e+02 -0.467 0.6458

q4 6.258e-05 2.815e-05 2.223 0.0393 *

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 6.247 on 18 degrees of freedom

Multiple R-squared: 0.9008, Adjusted R-squared: 0.8733

F-statistic: 32.71 on 5 and 18 DF, p-value: 1.986e-08

Приложение 6

Результаты регрессионного анализа по сегменту «программное обеспечение»

Модель 1. y ~ f1 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-10.223 -4.550 -0.068 4.594 15.932

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 26.263116 57.554251 0.456 0.6536

f1 3.306575 2.923580 1.131 0.2729

q1 -8.510367 10.119160 -0.841 0.4114

q2 0.048987 0.073992 0.662 0.5163

q3 138.091748 84.733540 1.630 0.1205

q4 0.006417 0.003339 1.922 0.0706 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.709 on 18 degrees of freedom

Multiple R-squared: 0.6591, Adjusted R-squared: 0.5644

F-statistic: 6.961 on 5 and 18 DF, p-value: 0.0008855

Модель 2. y ~ f2 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-9.3198 -4.3662 -0.9033 3.4927 9.2602

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) -2.272e+02 8.782e+01 -2.587 0.01859 *

f2 8.834e+00 2.499e+00 3.535 0.00237 **

q1 1.109e+01 1.003e+01 1.105 0.28352

q2 1.128e-02 5.995e-02 0.188 0.85286

q3 2.519e+02 7.195e+01 3.501 0.00255 **

q4 4.981e-03 2.663e-03 1.871 0.07776 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 6.129 on 18 degrees of freedom

Multiple R-squared: 0.7845, Adjusted R-squared: 0.7246

F-statistic: 13.1 on 5 and 18 DF, p-value: 1.791e-05

Модель 3. y ~ f3 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-9.3204 -5.7074 0.4094 4.8588 15.8410

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) -43.225010 69.174687 -0.625 0.5399

f3 5.787822 3.075807 1.882 0.0761 .

q1 -2.018405 10.484664 -0.193 0.8495

q2 0.039872 0.070271 0.567 0.5774

q3 177.716071 80.018376 2.221 0.0394 *

q4 0.006415 0.003152 2.036 0.0568 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.293 on 18 degrees of freedom

Multiple R-squared: 0.6949, Adjusted R-squared: 0.6102

F-statistic: 8.2 on 5 and 18 DF, p-value: 0.0003492

Модель 4. y ~ f5 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-10.8426 -4.4817 -0.2685 4.7844 16.6143

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 38.455458 60.950683 0.631 0.536

f5 0.093538 1.189655 0.079 0.938

q1 -10.882794 10.242298 -1.063 0.302

q2 0.053707 0.076562 0.701 0.492

q3 156.363660 90.976567 1.719 0.103

q4 0.006086 0.003442 1.768 0.094 .

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.977 on 18 degrees of freedom

Multiple R-squared: 0.635, Adjusted R-squared: 0.5336

F-statistic: 6.264 on 5 and 18 DF, p-value: 0.001561

Модель 5. y ~ f6 + q1 + q2 + q3 + q4

y ~ f6 + q1 + q2 + q3 + q4

Residuals:

Min 1Q Median 3Q Max

-11.5002 -4.9959 0.1463 4.3484 11.8248

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 80.448991 57.466129 1.400 0.1785

f6 -1.484425 0.789767 -1.880 0.0765 .

q1 -3.263343 10.207412 -0.320 0.7529

q2 0.063615 0.070127 0.907 0.3763

q3 138.132154 79.502800 1.737 0.0994 .

q4 0.005243 0.003179 1.649 0.1164

---

Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 7.294 on 18 degrees of freedom

Multiple R-squared: 0.6948, Adjusted R-squared: 0.61

F-statistic: 8.195 on 5 and 18 DF, p-value: 0.0003503

Приложение 7

Скрипт статистической модели в R Studio

replace_na <- function(data)

{

data[is.na(data)]<-0

return(data)

}

data_vector <- list()

number_of_companies = 16

for(i in 1:number_of_companies)

{

data_vector[[i]] <- data.frame(lapply(read.xlsx2(paste(as.character(i),".xlsx",sep = ''), sheetIndex = 1), function(x) as.numeric(as.character(x))))

}

segment_IS_EPS <- seq(0,0,length.out = 24)

segment_REVENUE_PER_SH <- seq(0,0,length.out = 24)

segment_EBITDA_PER_SH <- seq(0,0,length.out = 24)

segment_EV_TO_SH_OUT <- seq(0,0,length.out = 24)

segment_FREE_CASH_FLOW_PER_SH <- seq(0,0,length.out = 24)

segment_BOOK_VAL_PER_SH <- seq(0,0, length.out = 24)

segment_PX_LAST <- seq(0,0, length.out = 24)

segment_VISITORS <- seq(0,0, length.out = 24)

segment_BOUNCE_RATE <- seq(0,0, length.out = 24)

segment_DURATION <- seq(0,0, length.out = 24)

segment_PAGEVIEWS <- seq(0,0, length.out = 24)

for(j in 1:24)

{

for(i in 1:number_of_companies)

{

segment_IS_EPS[j] = segment_IS_EPS[j] + replace_na(data_vector[[i]]$IS_EPS[j])

segment_REVENUE_PER_SH[j] = segment_REVENUE_PER_SH[j] + replace_na(data_vector[[i]]$REVENUE_PER_SH[j])

segment_EBITDA_PER_SH[j] = segment_EBITDA_PER_SH[j] + replace_na(data_vector[[i]]$EBITDA_PER_SH[j])

segment_EV_TO_SH_OUT[j] = segment_EV_TO_SH_OUT[j] + replace_na(data_vector[[i]]$EV_TO_SH_OUT[j])

segment_FREE_CASH_FLOW_PER_SH[j] = segment_FREE_CASH_FLOW_PER_SH[j] + replace_na(data_vector[[i]]$FREE_CASH_FLOW_PER_SH[j])

segment_BOOK_VAL_PER_SH[j] = segment_BOOK_VAL_PER_SH[j] + replace_na(data_vector[[i]]$BOOK_VAL_PER_SH[j])

segment_PX_LAST[j] = segment_PX_LAST[j] + replace_na(data_vector[[i]]$PX_LAST[j])

segment_VISITORS[j] = segment_VISITORS[j] + replace_na(data_vector[[i]]$Visitors[j])

segment_BOUNCE_RATE[j] = segment_BOUNCE_RATE[j] + replace_na(data_vector[[i]]$Bouncerate[j])

segment_DURATION[j] = segment_DURATION[j] + replace_na(data_vector[[i]]$Duration[j])

segment_PAGEVIEWS[j] = segment_PAGEVIEWS[j] + replace_na(data_vector[[i]]$Pageviews[j])

}

}

final_data <- data.frame(segment_PAGEVIEWS,segment_DURATION,segment_BOUNCE_RATE,segment_VISITORS/segment_REVENUE_PER_SH,segment_IS_EPS,segment_REVENUE_PER_SH,segment_EBITDA_PER_SH,

segment_EV_TO_SH_OUT,segment_FREE_CASH_FLOW_PER_SH,

segment_BOOK_VAL_PER_SH,segment_PX_LAST)

names(final_data) <- c("PAGEVIEWS","DURATION","BOUNCE_RATE","VISITORS","IS_EPS", "REVENUE_PER_SH",

"EBITDA_PER_SH","EV_TO_SH_OUT","FREE_CASH_FLOW_PER_SH",

"BOOK_VAL_PER_SH","PX_LAST")

final_data <- final_data/number_of_companies

sink(file = "content.txt")

summary(final_data)

#name independent variables

#f - financial, q - quantative, y - dependent variable

y = final_data$PX_LAST

f1 = final_data$IS_EPS

f2 = final_data$REVENUE_PER_SH

f3 = final_data$EBITDA_PER_SH

f4 = final_data$EV_TO_SH_OUT #exclude

f5 = final_data$FREE_CASH_FLOW_PER_SH

f6 = final_data$BOOK_VAL_PER_SH

q1 = final_data$PAGEVIEWS

q2 = final_data$DURATION

q3 = final_data$BOUNCE_RATE

q4 = final_data$VISITORS

fit <- lm(y ~ f5 + q1 + q2 + q3 + q4)

summary(fit)

plot(fit$fit,fit$res,xlab="Fitted",ylab="Residuals", main="Residual-Fitted plot") #there you expect to see random

abline(h=0)

qqnorm(fit$res) #expect to see y=x line

Приложение 8

Скрипт алгоритма для выгрузки операционных данных с ресурса alexa.com

import csv

import urllib

import yaml

DATES = "start=01-2014&end=12-2014"

KEY = "4d7f98a965ae01204f65a80a5a247f21"

BASE = "http://api.similarweb.com/Site/"

VPARAMS = "/v1/visits?" + DATES + "&Format=JSON&UserKey=4d7f98a965ae01204f65a80a5a247f21"

PAGEPARAMS = "/v1/pageviews?" + DATES + "&Format=JSON&UserKey=4d7f98a965ae01204f65a80a5a247f21"

DURATIONPARAMS = "/v1/visitduration?" + DATES + "&Format=JSON&UserKey=4d7f98a965ae01204f65a80a5a247f21"

BOUNCERATEPARAMS = "/v1/bouncerate?" + DATES + "&Format=JSON&UserKey=4d7f98a965ae01204f65a80a5a247f21"

COUNTRYPARAMS = "/v1/traffic?Format=JSON&UserKey=" + KEY

def get_data(URL):

print URL

visits = []

dates = []

pageviews = []

duration = []

bouncerate = []

visits_url = BASE + URL + VPARAMS

pageviews_url = BASE + URL +PAGEPARAMS

duration_url = BASE + URL + DURATIONPARAMS

bouncerate_url = BASE + URL + BOUNCERATEPARAMS

res = urllib.urlopen(visits_url).read()

traffic = yaml.safe_load(res)

print traffic

for x in traffic.itervalues():

for z in x:

visits.append(z['Value'])

dates.append(z['Date'])

res_pageviews = urllib.urlopen(pageviews_url).read()

pv = yaml.safe_load(res_pageviews)

for x in pv.itervalues():

for z in x:

pageviews.append(z['Value'])

res_duration = urllib.urlopen(duration_url).read()

dr = yaml.safe_load(res_duration)

for x in dr.itervalues():

for z in x:

duration.append(z['Value'])

res_bounce = urllib.urlopen(bouncerate_url).read()

br = yaml.safe_load(res_bounce)

for x in br.itervalues():

for z in x:

bouncerate.append(z['Value'])

with open(URL + DATES + '.csv', 'wb') as f:

w = csv.writer(f)

w.writerow(["Date", "Visitors", "Pageviews", "Duration", "Bouncerate"])

for i in range(len(visits)):

w.writerow([dates[i], visits[i], pageviews[i], duration[i], bouncerate[i]])

return traffic

def get_urls(file):

company_url = []

with open(file) as urls:

for url in urls:

company_url.append(url[:-1])

return company_url

urls = get_urls("list.txt")

print(get_data(urls[1]))

#get all data for given date

#uncommnet if have an paid api

#for url in urls:

#get_data(url)

Размещено на Allbest.ru

...