Оценка результатов построения деревьев решений при помощи регрессионного анализа

39) age=35-44,25-34,45-54,55-64 1494 580 Happy (0.6117805 0.3882195)

78) Marital=Married 1334 492 Happy (0.6311844 0.3688156) *

79) Marital=Single 160 72 Unhappy (0.4500000 0.5500000)

158) Real_inc>=12613.96 74 30 Happy (0.5945946 0.4054054) *

159) Real_inc< 12613.96 86 28 Unhappy (0.3255814 0.6744186) *

5) Marital=No longer married 319 114 Unhappy (0.3573668 0.6426332) *

3) Health=Fair health,Bad health 4748 1755 Unhappy (0.3696293 0.6303707)

6) Marital=Married 3389 1444 Unhappy (0.4260844 0.5739156)

12) age=18-24,25-34 695 319 Happy (0.5410072 0.4589928)

24) status=Countryside,Small town 151 54 Happy (0.6423841 0.3576159) *

25) status=Town,Regional center 544 265 Happy (0.5128676 0.4871324)

50) Real_inc< 836.5103 62 21 Happy (0.6612903 0.3387097) *

51) Real_inc>=836.5103 482 238 Unhappy (0.4937759 0.5062241)

102) Real_inc>=19838.7 147 63 Happy (0.5714286 0.4285714) *

103) Real_inc< 19838.7 335 154 Unhappy (0.4597015 0.5402985)

206) Own_home=Do not own house 61 25 Happy (0.5901639 0.4098361) *

207) Own_home=Own house 274 118 Unhappy (0.4306569 0.5693431)

414) Real_inc< 7582.569 91 44 Happy (0.5164835 0.4835165)

828) Real_inc>=6061.862 30 10 Happy (0.6666667 0.3333333) *

829) Real_inc< 6061.862 61 27 Unhappy (0.4426230 0.5573770) *

415) Real_inc>=7582.569 183 71 Unhappy (0.3879781 0.6120219) *

13) age=35-44,45-54,55-64 2694 1068 Unhappy (0.3964365 0.6035635)

26) Real_inc>=19025 664 328 Unhappy (0.4939759 0.5060241)

52) status=Town,Small town 243 98 Happy (0.5967078 0.4032922) *

53) status=Countryside,Regional center 421 183 Unhappy (0.4346793 0.5653207) *

27) Real_inc< 19025 2030 740 Unhappy (0.3645320 0.6354680) *

7) Marital=Single,No longer married 1359 311 Unhappy (0.2288447 0.7711553) *

Спецификация дерева CHAID

Model formula:

Unhapp_k_3 ~ sex + status + age + educ + Rich + Poor + Rel_rich +

Rel_poor + Health + Obese + Overweight + Employment + Is_russian +

Marital + Has_children + Own_home + Savings

Fitted party:

[1] root

| [2] Health in Fair health

| | [3] Marital in Single: Unhappy (n = 441, err = 29.9%)

| | [4] Marital in Married

| | | [5] age in 35-44

| | | | [6] Has_children in No children: Unhappy (n = 58, err = 22.4%)

| | | | [7] Has_children in Has children

| | | | | [8] Is_russian in Other: Happy (n = 90, err = 37.8%)

| | | | | [9] Is_russian in Russian_nationality

| | | | | | [10] Rel_rich in Not relatively rich: Unhappy (n = 542, err = 41.7%)

| | | | | | [11] Rel_rich in Relatively_rich: Happy (n = 95, err = 41.1%)

| | | [12] age in 18-24, 25-34

| | | | [13] status in Town, Regional center: Happy (n = 488, err = 48.6%)

| | | | [14] status in Countryside, Small town: Happy (n = 143, err = 34.3%)

| | | [15] age in 45-54, 55-64

| | | | [16] Poor in Not poor

| | | | | [17] status in Town, Countryside, Small town

| | | | | | [18] Own_home in Do not own house: Unhappy (n = 21, err = 23.8%)

| | | | | | [19] Own_home in Own house: Unhappy (n = 651, err = 45.8%)

| | | | | [20] status in Regional center: Unhappy (n = 490, err = 34.9%)

| | | | [21] Poor in Poor: Unhappy (n = 333, err = 31.5%)

| | [22] Marital in No longer married

| | | [23] Has_children in No children: Unhappy (n = 61, err = 8.2%)

| | | [24] Has_children in Has children

| | | | [25] Rel_rich in Not relatively rich: Unhappy (n = 440, err = 20.0%)

| | | | [26] Rel_rich in Relatively_rich: Unhappy (n = 150, err = 30.7%)

| [27] Health in Bad health: Unhappy (n = 484, err = 25.0%)

| [28] Health in Good health

| | [29] Marital in Single

| | | [30] Is_russian in Other: Happy (n = 199, err = 16.6%)

| | | [31] Is_russian in Russian_nationality

| | | | [32] age in 35-44, 45-54: Unhappy (n = 79, err = 40.5%)

| | | | [33] age in 18-24, 55-64: Happy (n = 409, err = 34.0%)

| | | | [34] age in 25-34: Happy (n = 185, err = 47.6%)

| | [35] Marital in Married

| | | [36] Is_russian in Other: Happy (n = 391, err = 18.4%)

| | | [37] Is_russian in Russian_nationality

| | | | [38] educ in Middle, Low

| | | | | [39] age in 35-44, 45-54, 55-64

| | | | | | [40] Savings in No savings

| | | | | | | [41] status in Town, Countryside, Regional center: Happy (n = 604, err = 42.5%)

| | | | | | | [42] status in Small town: Happy (n = 51, err = 21.6%)

| | | | | | [43] Savings in Has savings: Happy (n = 92, err = 25.0%)

| | | | | [44] age in 18-24: Happy (n = 109, err = 16.5%)

| | | | | [45] age in 25-34

| | | | | | [46] Obese in Not obese: Happy (n = 462, err = 36.6%)

| | | | | | [47] Obese in Obese: Happy (n = 52, err = 15.4%)

| | | | [48] educ in Higher: Happy (n = 784, err = 24.6%)

| | [49] Marital in No longer married: Unhappy (n = 304, err = 35.2%)

Number of inner nodes: 21

Number of terminal nodes: 28

Спецификация дерева C50

Call:

C5.0.formula(formula = Unhapp_k_3 ~., data = train)

Class specified by attribute `outcome'

Read 8679 cases (19 attributes) from undefined.data

Decision tree:

Health = Good health:

:...Marital = No longer married: Unhappy (323.7/114.9)

: Marital in {Single,Married}:

: :...Is_russian = Other: Happy (626.6/109)

: Is_russian = Russian_nationality:

: :...Marital = Married: Happy (2283.6/725.3)

: Marital = Single:

: :...age in {35-44,45-54,55-64}: Unhappy (96.6/40.7)

: age = 18-24: Happy (418/140.6)

: age = 25-34:

: :...Rel_poor = Not relatively poor: Happy (156.9/66.4)

: Rel_poor = Relatively_poor: Unhappy (41/15)

Health in {Fair health,Bad health}:

:...Marital in {Single,No longer married}: Unhappy (1358.9/311.9)

Marital = Married:

:...age in {18-24,25-34}:

:...status = Small town: Happy (25.5/10.5)

: status = Countryside:

: :...Health = Fair health: Happy (123.5/41.5)

: : Health = Bad health: Unhappy (3)

: status = Town:

: :...Rich = Not rich:

: : :...age = 18-24: Happy (22/7)

: : : age = 25-34: Unhappy (147.4/58)

: : Rich = Rich:

: : :...age = 18-24: Unhappy (3)

: : age = 25-34: Happy (41.4/12.4)

: status = Regional center:

: :...Own_home = Do not own house: Happy (64/22)

: Own_home = Own house:

: :...educ = Higher: Happy (152.1/66.5)

: educ = Low: Unhappy (18/6)

: educ = Middle:

: :...Rel_poor = Not relatively poor: Unhappy (75/28)

: Rel_poor = Relatively_poor: Happy (22/6)

age in {35-44,45-54,55-64}:

:...Rel_rich = Relatively_rich:

:...Is_russian = Other: Happy (52/17)

: Is_russian = Russian_nationality:

: :...status in {Town,Small town}: Happy (122.6/52.1)

: status = Countryside: Unhappy (63/29)

: status = Regional center:

: :...age in {45-54,55-64}: Unhappy (109.3/39.3)

: age = 35-44:

: :...Health = Bad health: Unhappy (3)

: Health = Fair health:

: :...Has_children = No children: Unhappy (7/1)

: Has_children = Has children: Happy (48.5/20)

Rel_rich = Not relatively rich:

:...Has_children = No children: Unhappy (138.1/35)

Has_children = Has children:

:...Own_home = Do not own house: Unhappy (119.7/35)

Own_home = Own house:

:...Health = Bad health:

:...educ in {Middle,Low}: Unhappy (177.4/45.2)

: educ = Higher:

: :...Poor = Poor: Happy (7.1/1.1)

: Poor = Not poor:

: :...Savings = No savings: Unhappy (28.3/9.1)

: Savings = Has savings: Happy (6.2/1.2)

Health = Fair health:

:...Is_russian = Other:

:...age = 35-44: Happy (76.5/29.5)

: age in {45-54,55-64}: Unhappy (133.5/56)

Is_russian = Russian_nationality:

:...Savings = Has savings:

:...Overweight = Not overweight: Happy (124.3/56.8)

: Overweight = Overweight: Unhappy (75.8/25.4)

Savings = No savings:

:...Employment in {Disabled,Inactive,

: Unemployed}: Unhappy (160.1/53.1)

Employment = Retired:

:...educ = Low: Unhappy (43.5/14.5)

: educ = Middle: [S1]

: educ = Higher: [S2]

Employment = Employed:

:...age = 45-54: Unhappy (353.7/117.9)

age in {35-44,55-64}:

:...Real_inc <= 4368.257:

:...Obese = Obese: Happy (10.2)

: Obese = Not obese:

: :...sex = Male: Unhappy (9.8/1.8)

: sex = Female: Happy (21.9/6.9)

Real_inc > 4368.257:

:...Poor = Poor: Unhappy (53.5/15)

Poor = Not poor: [S3]

Спецификация дерева C4.5/J48

J48 pruned tree

------------------

Health = Fair health

| Marital = Single: Unhappy (432.0/128.0)

| Marital = Married

| | Rel_rich = Not relatively rich

| | | age = 35-44

| | | | Is_russian = Other: Happy (84.0/33.0)

| | | | Is_russian = Russian_nationality: Unhappy (589.0/236.0)

| | | age = 18-24: Happy (53.0/19.0)

| | | age = 25-34

| | | | status = Town: Unhappy (137.0/58.0)

| | | | status = Countryside: Happy (100.0/35.0)

| | | | status = Regional center

| | | | | sex = Male: Unhappy (94.0/44.0)

| | | | | sex = Female: Happy (104.0/44.0)

| | | | status = Small town: Happy (15.0/7.0)

| | | age = 45-54: Unhappy (588.0/208.0)

| | | age = 55-64: Unhappy (661.0/259.0)

| | Rel_rich = Relatively_rich

| | | status = Town: Happy (120.0/44.0)

| | | status = Countryside: Happy (67.0/29.0)

| | | status = Regional center: Unhappy (190.0/87.0)

| | | status = Small town: Happy (15.0/4.0)

| Marital = No longer married: Unhappy (646.0/139.0)

Health = Bad health: Unhappy (477.0/118.0)

Health = Good health

| Marital = Single

| | Is_russian = Other: Happy (191.0/33.0)

| | Is_russian = Russian_nationality

| | | age = 35-44: Unhappy (62.0/25.0)

| | | age = 18-24: Happy (387.0/131.0)

| | | age = 25-34

| | | | Real_inc <= 13951.216793: Unhappy (87.0/32.0)

| | | | Real_inc > 13951.216793: Happy (94.0/32.0)

| | | age = 45-54: Unhappy (16.0/7.0)

| | | age = 55-64: Unhappy (9.0/4.0)

| Marital = Married: Happy (2470.0/733.0)

| Marital = No longer married: Unhappy (298.0/103.0)

Number of Leaves : 26

Size of the tree: 38

Спецификация дерева LMT

Logistic model tree

------------------

Health = Fair health

| Marital = Single: LM_1:1/3 (432)

| Marital = Married

| | Rel_rich = Not relatively rich: LM_2:1/4 (2425)

| | Rel_rich = Relatively_rich

| | | Is_russian = Other: LM_3:1/5 (47)

| | | Is_russian = Russian_nationality: LM_4:1/5 (345)

| Marital = No longer married: LM_5:1/3 (646)

Health = Bad health: LM_6:1/2 (477)

Health = Good health: LM_7:1/2 (3614)

Number of Leaves : 7

Size of the Tree: 11

«Divergent drinking patterns and factors affecting homemade alcohol consumption (the case of Russia)»

Таблица 8

Спецификации регрессии

	Samogon_consumption
Sex1	-0.974***
age	0.016**
College_ed1	-0.015
College_ed3	0.453**
Employment1	-0.861***
Ln_income	-0.023
South1	0.259
Vilage1	0.938***
Constant	-2.901***
N	3,025
Log Likelihood	-692.437
AIC	1,402.874
*p <.05; **p <.01; ***p <.001

AUC_binomial

0.7011606

cutoff Accuracy Sensitivity Specificity

1: 0.5 0.5441459 0.8219178 0.5232198

Спецификация дерева Rpart

n= 3124

node), split, n, loss, yval, (yprob)

* denotes terminal node

1) root 3124 223 0 (0.92861716 0.07138284) *

> AUC_rpart

[1] 0.5

> measures_rpart

cutoff Accuracy Sensitivity Specificity

1: 0.5 0.92994242 0 1

Спецификация дерева CHAID

Fitted party:

[1] root

| [2] Vilage in 0

| | [3] Employment in 0

| | | [4] Sex in 0

| | | | [5] South in 0: 0 (n = 341, err = 12.0%)

| | | | [6] South in 1

| | | | | [7] College_ed in 0, 1: 0 (n = 18, err = 5.6%)

| | | | | [8] College_ed in 3: 1 (n = 18, err = 38.9%)

| | | [9] Sex in 1: 0 (n = 432, err = 4.4%)

| | [10] Employment in 1

| | | [11] Sex in 0: 0 (n = 901, err = 4.8%)

| | | [12] Sex in 1: 0 (n = 743, err = 1.7%)

| [13] Vilage in 1

| | [14] Employment in 0

| | | [15] College_ed in 0, 1: 0 (n = 70, err = 10.0%)

| | | [16] College_ed in 3: 0 (n = 189, err = 27.5%)

| | [17] Employment in 1

| | | [18] College_ed in 0, 1

| | | | [19] Sex in 0: 0 (n = 91, err = 5.5%)

| | | | [20] Sex in 1: 0 (n = 73, err = 0.0%)

| | | [21] College_ed in 3: 0 (n = 241, err = 12.9%)

Number of inner nodes: 10

Number of terminal nodes: 11

AUC_ch

[1] 0.7171565

measures_ch

cutoff Accuracy Sensitivity Specificity

1: 1.00000000 0.92994242 0.0000000 1.00000000

2: 0.61111111 0.92418426 0.0000000 0.99380805

3: 0.27513228 0.89347409 0.2191781 0.94427245

4: 0.12863071 0.83109405 0.3561644 0.86687307

5: 0.12023460 0.75239923 0.5753425 0.76573787

6: 0.10000000 0.71785029 0.5890411 0.72755418

7: 0.05555556 0.71497121 0.5890411 0.72445820

8: 0.05494505 0.68714012 0.6301370 0.69143447

9: 0.04772475 0.44049904 0.8630137 0.40866873

10: 0.04398148 0.31094050 0.9589041 0.26212590

11: 0.01749664 0.08733205 0.9726027 0.02063983

12: 0.00000000 0.07005758 1.0000000 0.00000000

13: 0.00000000 0.07005758 1.0000000 0.00000000

Спецификация дерева C50

Read 3124 cases (8 attributes) from undefined.data

Decision tree:

0 (3124/223)

Evaluation on training data (3124 cases):

2901 (a): class 0

223 (b): class 1

> AUC_c50

[1] 0.5

measures50

cutoff Accuracy Sensitivity Specificity

1: 0.5 0.92994242 0 1

Спецификация дерева C4.5/J48

J48 pruned tree

------------------

: 0 (3025.0/217.0)

Number of Leaves : 1

Size of the tree: 1

> AUC_j48

[1] 0.5

> measures_j48

cutoff Accuracy Sensitivity Specificity

1: 0.5 0.92994242 0 1

Спецификация дерева LMT

Logistic model tree

------------------

: LM_1:1/1 (3025)

Number of Leaves : 1

Size of the Tree: 1

LM_1:

Class 0:

0.9 +

[Vilage=1] * -0.19

Class 1:

-0.9 +

[Vilage=1] * 0.19

> AUC_LMT

[1] 0.609787

> measures_lmt

cutoff Accuracy Sensitivity Specificity

1: 1.0000000 0.92994242 0.0000000 1.0000000

2: 0.1944205 0.75719770 0.4383562 0.7812178

3: 0.1427246 0.07005758 1.0000000 0.0000000

4: 0.0000000 0.07005758 1.0000000 0.0000000

«Низкооплачиваемые рабочие места на российском рынке труда: есть ли выход и куда он ведет?»

Таблица 9

Спецификации регрессии

	TARGET
educHigher	-1.135***
educProfessional	-0.636***
Age35-44	-0.353**
Age45-54	-0.287*
Age55-64	0.402*
City_sizeCity	1.250***
City_sizeSmall_town_or_village	1.945***
City_sizeTown	1.649***
sectorAgriculture	1.188***
sectorCommercial_services	0.319**
sectorConstruction	-0.187
sectorNon-comercial_services	1.122***
Professional_statusBlue_collar_qual	0.184
Professional_statusBlue_collar_unqal	1.168***
Professional_statusWhite_collar_unqual	1.088***
Male	-0.951***
Married	0.101
Children_under_18	0.010
Full_day	0.646***
Constant	-3.208***
N	4,304
Log Likelihood	-2,014.644
AIC	4,069.287
*p <.05; **p <.01; ***p <.001

Спецификация дерева Rpart

n= 5675

node), split, n, loss, yval, (yprob)

* denotes terminal node

1) root 5675 1454 0 (0.7437885 0.2562115)

2) Professional_status=White_collar_qual,Blue_collar_qual 3983 704 0 (0.8232488 0.1767512)

4) sector=Industry,Commercial_services,Construction 2635 329 0 (0.8751423 0.1248577) *

5) sector=Agriculture,Non-comercial_services 1348 375 0 (0.7218101 0.2781899)

10) educ=Higher 704 116 0 (0.8352273 0.1647727) *

11) educ=Secondary or lower,Professional 644 259 0 (0.5978261 0.4021739)

22) City_size=Moskow_or_spb,City 169 40 0 (0.7633136 0.2366864) *

23) City_size=Small_town_or_village,Town 475 219 0 (0.5389474 0.4610526)

46) City_size=Town 168 67 0 (0.6011905 0.3988095)

92) Male>=0.5 40 10 0 (0.7500000 0.2500000) *

93) Male< 0.5 128 57 0 (0.5546875 0.4453125)

186) Age=35-44,45-54 75 28 0 (0.6266667 0.3733333) *

187) Age=24-34,55-64 53 24 1 (0.4528302 0.5471698) *

47) City_size=Small_town_or_village 307 152 0 (0.5048860 0.4951140)

94) Age=24-34,35-44,45-54 251 119 0 (0.5258964 0.4741036)

188) Age=24-34 63 26 0 (0.5873016 0.4126984) *

189) Age=35-44,45-54 188 93 0 (0.5053191 0.4946809)

378) Children_under_18< 0.5 77 34 0 (0.5584416 0.4415584) *

379) Children_under_18>=0.5 111 52 1 (0.4684685 0.5315315)

758) educ=Secondary or lower 55 25 0 (0.5454545 0.4545455) *

759) educ=Professional 56 22 1 (0.3928571 0.6071429) *

95) Age=55-64 56 23 1 (0.4107143 0.5892857) *

3) Professional_status=Blue_collar_unqal,White_collar_unqual 1692 750 0 (0.5567376 0.4432624)

6) City_size=Moskow_or_spb 170 22 0 (0.8705882 0.1294118) *

7) City_size=City,Small_town_or_village,Town 1522 728 0 (0.5216820 0.4783180)

14) Age=24-34,35-44,45-54 1311 580 0 (0.5575896 0.4424104)

28) sector=Industry,Commercial_services,Construction 942 368 0 (0.6093418 0.3906582)

56) City_size=City 356 105 0 (0.7050562 0.2949438) *

57) City_size=Small_town_or_village,Town 586 263 0 (0.5511945 0.4488055)

114) educ=Higher,Professional 261 90 0 (0.6551724 0.3448276) *

115) educ=Secondary or lower 325 152 1 (0.4676923 0.5323077)

230) Male>=0.5 105 39 0 (0.6285714 0.3714286) *

231) Male< 0.5 220 86 1 (0.3909091 0.6090909)

462) City_size=Town 113 50 1 (0.4424779 0.5575221)

924) Age=35-44 34 15 0 (0.5588235 0.4411765) *

925) Age=24-34,45-54 79 31 1 (0.3924051 0.6075949) *

463) City_size=Small_town_or_village 107 36 1 (0.3364486 0.6635514) *

29) sector=Agriculture,Non-comercial_services 369 157 1 (0.4254743 0.5745257)

58) Male>=0.5 149 67 0 (0.5503356 0.4496644)

116) Age=24-34,35-44 117 49 0 (0.5811966 0.4188034)

232) City_size=City,Town 65 22 0 (0.6615385 0.3384615) *

233) City_size=Small_town_or_village 52 25 1 (0.4807692 0.5192308) *

117) Age=45-54 32 14 1 (0.4375000 0.5625000) *

59) Male< 0.5 220 75 1 (0.3409091 0.6590909)

118) educ=Higher 32 12 0 (0.6250000 0.3750000) *

119) educ=Secondary or lower,Professional 188 55 1 (0.2925532 0.7074468) *

15) Age=55-64 211 63 1 (0.2985782 0.7014218) *

Спецификация дерева CHAID

Model formula:

TARGET ~ educ + Age + City_size + sector + Professional_status +

Male + Married + Children_under_18 + Full_day

Fitted party:

[1] root

| [2] Professional_status in White_collar_qual, Blue_collar_qual

| | [3] sector in Industry, Construction

| | | [4] educ in Secondary or lower: 0 (n = 437, err = 15.3%)

| | | [5] educ in Higher, Professional

| | | | [6] Professional_status in White_collar_qual, Blue_collar_unqal, White_collar_unqual: 0 (n = 354, err = 4.8%)

| | | | [7] Professional_status in Blue_collar_qual: 0 (n = 221, err = 10.0%)

| | [8] sector in Agriculture: 0 (n = 122, err = 41.0%)

| | [9] sector in Commercial_services

| | | [10] City_size in Moskow_or_spb: 0 (n = 131, err = 5.3%)

| | | [11] City_size in City: 0 (n = 372, err = 12.1%)

| | | [12] City_size in Small_town_or_village, Town: 0 (n = 490, err = 19.8%)

| | [13] sector in Non-comercial_services

| | | [14] educ in Secondary or lower: 1 (n = 114, err = 49.1%)

| | | [15] educ in Higher

| | | | [16] Age in 24-34: 0 (n = 119, err = 35.3%)

| | | | [17] Age in 35-44: 0 (n = 216, err = 14.8%)

| | | | [18] Age in 45-54, 55-64: 0 (n = 198, err = 7.1%)

| | | [19] educ in Professional: 0 (n = 303, err = 35.6%)

| [20] Professional_status in Blue_collar_unqal, White_collar_unqual

| | [21] City_size in Moskow_or_spb: 0 (n = 110, err = 14.5%)

| | [22] City_size in City: 0 (n = 377, err = 38.5%)

| | [23] City_size in Small_town_or_village: 1 (n = 394, err = 41.4%)

| | [24] City_size in Town: 0 (n = 346, err = 49.1%)

Number of inner nodes: 8

Number of terminal nodes: 16

Спецификация дерева C50

C5.0.formula(formula = TARGET ~., data = train)

Class specified by attribute `outcome'

Read 5675 cases (10 attributes) from undefined.data

Decision tree:

Professional_status in {White_collar_qual,Blue_collar_qual}:

:...sector in {Industry,Commercial_services,Construction}: 0 (2629.7/329)

: sector in {Agriculture,Non-comercial_services}:

: :...educ = Higher: 0 (704.8/116)

: educ in {Secondary or lower,Professional}:

: :...City_size in {Moskow_or_spb,City,Town}: 0 (339.4/107.3)

: City_size = Small_town_or_village:

: :...Age = 24-34: 0 (67/28.2)

: Age = 35-44:

: :...Children_under_18 <= 0: 0 (22.3/6.4)

: : Children_under_18 > 0: 1 (74.6/33.2)

: Age = 45-54:

: :...Married <= 0: 0 (32.2/11.4)

: : Married > 0: 1 (51.4/22.1)

: Age = 55-64:

: :...Married <= 0: 1 (18.3/3.5)

: Married > 0:

: :...Male <= 0: 0 (21.6/7)

: Male > 0: 1 (19.7/6.5)

Professional_status in {Blue_collar_unqal,White_collar_unqual}:

:...City_size = Moskow_or_spb: 0 (170.3/22)

City_size in {City,Small_town_or_village,Town}:

:...Age = 55-64: 1 (235.3/74.3)

Age in {24-34,35-44,45-54}:

:...educ in {Higher,Professional}: 0 (589.7/206.6)

educ = Secondary or lower:

:...Male > 0:

:...Children_under_18 <= 0: 1 (55.7/24.8)

: Children_under_18 > 0: 0 (185.8/59.6)

Male <= 0:

:...sector = Construction: 0 (1)

sector = Non-comercial_services: 1 (113.8/27.9)

sector in {Industry,Agriculture,Commercial_services}:

:...City_size = City: 0 (108.6/40.7)

City_size in {Small_town_or_village,

Town}: 1 (233.9/90.5)

Спецификация дерева C4.5/J48

J48 pruned tree

------------------

Professional_status = White_collar_qual: 0 (1932.0/358.0)

Professional_status = Blue_collar_qual: 0 (1145.0/201.0)

Professional_status = Blue_collar_unqal

| City_size = Moskow_or_spb: 0 (29.0/2.0)

| City_size = City: 0 (72.0/26.0)

| City_size = Small_town_or_village: 1 (93.0/37.0)

| City_size = Town: 1 (77.0/31.0)

Professional_status = White_collar_unqual

| City_size = Moskow_or_spb: 0 (81.0/14.0)

| City_size = City

| | Age = 24-34: 0 (94.0/38.0)

| | Age = 35-44: 0 (84.0/21.0)

| | Age = 45-54: 0 (72.0/25.0)

| | Age = 55-64: 1 (55.0/20.0)

| City_size = Small_town_or_village

| | educ = Secondary or lower: 1 (182.0/60.0)

| | educ = Higher: 0 (39.0/11.0)

| | educ = Professional: 1 (80.0/38.0)

| City_size = Town

| | Age = 24-34: 0 (70.0/32.0)

| | Age = 35-44: 0 (96.0/30.0)

| | Age = 45-54: 1 (70.0/34.0)

| | Age = 55-64: 1 (33.0/7.0)

Number of Leaves : 18

Size of the tree: 24

Спецификация дерева LMT

Logistic model tree

------------------

Professional_status = White_collar_qual: LM_1:1/2 (1932)

Professional_status = Blue_collar_qual

| City_size = Moskow_or_spb: LM_2:1/3 (92)

| City_size = City: LM_3:1/3 (333)

| City_size = Small_town_or_village: LM_4:1/3 (335)

| City_size = Town

| | Male <= 0: LM_5:1/4 (68)

| | Male > 0: LM_6:1/4 (317)

Professional_status = Blue_collar_unqal

| City_size = Moskow_or_spb: LM_7:1/3 (29)

| City_size = City

| | Male <= 0: LM_8:1/4 (44)

| | Male > 0: LM_9:1/4 (28)

| City_size = Small_town_or_village: LM_10:1/3 (93)

| City_size = Town: LM_11:1/3 (77)

Professional_status = White_collar_unqual

| City_size = Moskow_or_spb: LM_12:1/3 (81)

| City_size = City

| | Age = 24-34

| | | Male <= 0: LM_13:1/5 (69)

| | | Male > 0: LM_14:1/5 (25)

| | Age = 35-44: LM_15:1/4 (84)

| | Age = 45-54: LM_16:1/4 (72)

| | Age = 55-64: LM_17:1/4 (55)

| City_size = Small_town_or_village

| | educ = Secondary or lower: LM_18:1/4 (182)

| | educ = Higher: LM_19:1/4 (39)

| | educ = Professional: LM_20:1/4 (80)

| City_size = Town

| | Age = 24-34: LM_21:1/4 (70)

| | Age = 35-44: LM_22:1/4 (96)

| | Age = 45-54: LM_23:1/4 (70)

| | Age = 55-64: LM_24:1/4 (33)

Number of Leaves : 24

Size of the Tree: 34

Категориальная небинарная целевая переменная

«Дороги, ведущие молодежь в NEET: случай России»



Таблица 10

Спецификация регрессионной модели

	Employed	NEET-inactive	NEET-unemployed
	(1)	(2)	(3)
statusCity	0.804**	0.366	0.445
statusMoscow	0.917*	1.503**	1.448
statusTown or village	0.353	1.129**	-0.326
Health_statusGood health	0.179	0.439	0.269
Ln_inc	0.125***	0.092*	-0.135*
age15-19	-1.821***	-2.523***	-3.325***
educationCollege or higher	-0.759	-1.344**	-1.315
educationComplete middle	0.272	1.023*	1.164
educationComplete middle plus tech school	0.142	0.066	0.390
educationIncomplete middle plus tech school	0.992	1.681*	1.684
educationProfessional middle	2.030**	1.281	2.288*
marriedMarried	0.693	-0.248	0.309
Constant	-0.464	-1.055	-0.677
AIC	1,125.378	1,125.378	1,125.378
*p <.05; **p <.01; ***p <.001

Спецификация дерева Rpart

n= 678

node), split, n, loss, yval, (yprob)

* denotes terminal node

1) root 678 367 Students (0.45870206 0.34365782 0.15191740 0.04572271)

2) age=15-19 342 97 Students (0.71637427 0.18128655 0.08771930 0.01461988) *

3) age=20-24 336 165 Employed (0.19642857 0.50892857 0.21726190 0.07738095)

6) Ln_inc>=7.523116 236 88 Employed (0.12288136 0.62711864 0.22033898 0.02966102)

12) status=Large city,City,Moscow 157 50 Employed (0.15923567 0.68152866 0.12738854 0.03184713) *

13) status=Town or village 79 38 Employed (0.05063291 0.51898734 0.40506329 0.02531646)

26) education=College or higher,Professional middle 36 14 Employed (0.08333333 0.61111111 0.25000000 0.05555556) *

27) education=Incomplete middle,Complete middle,Complete middle plus tech school 43 20 NEET-inactive (0.02325581 0.44186047 0.53488372 0.00000000) *

7) Ln_inc< 7.523116 100 63 Students (0.37000000 0.23000000 0.21000000 0.19000000)

14) education=College or higher 52 23 Students (0.55769231 0.19230769 0.11538462 0.13461538) *

15) education=Incomplete middle,Complete middle,Complete middle plus tech school,Incomplete middle plus tech school,Professional middle 48 33 NEET-inactive (0.16666667 0.27083333 0.31250000 0.25000000) *

Спецификация дерева CHAID

Model formula:

Ocupation ~ status + Health_status + age + education + married

Fitted party:

[1] root

| [2] age in 20-24

| | [3] education in Incomplete middle, Complete middle, Complete middle plus tech school, Incomplete middle plus tech school, Professional middle

| | | [4] Health_status in Bad health: Employed (n = 33, err = 54.5%)

| | | [5] Health_status in Good health

| | | | [6] education in Incomplete middle, College or higher, Complete middle, Incomplete middle plus tech school: NEET-inactive (n = 50, err = 48.0%)

| | | | [7] education in Complete middle plus tech school, Professional middle

| | | | | [8] married in Not married: Employed (n = 80, err = 42.5%)

| | | | | [9] married in Married: Employed (n = 26, err = 11.5%)

| | [10] education in College or higher: Employed (n = 135, err = 54.8%)

| [11] age in 15-19

| | [12] education in Incomplete middle, College or higher, Complete middle plus tech school: Students (n = 218, err = 24.8%)

| | [13] education in Complete middle, Incomplete middle plus tech school: Students (n = 50, err = 52.0%)

| | [14] education in Professional middle: Employed (n = 3, err = 0.0%)

Number of inner nodes: 6

Number of terminal nodes: 8

Спецификация дерева C50

Call:

C5.0.formula(formula = Ocupation ~., data = train)

C5.0 [Release 2.07 GPL Edition]

-------------------------------

Class specified by attribute `outcome'

Read 678 cases (7 attributes) from undefined.data

Decision tree:

age = 15-19: Students (342/97)

age = 20-24:

:...Ln_inc > 7.467942: Employed (234.6/88.4)

Ln_inc <= 7.467942:

:...education in {Incomplete middle,

: Incomplete middle plus tech school}: NEET-inactive (8.1/3.1)

education in {College or higher,

: Complete middle plus tech school}: Students (66.5/33.5)

education in {Complete middle,

Professional middle}: Employed (26.9/16.6)

1: 0.6681416 0.3948913 0.8571129

Спецификация дерева C4.5/J48

J48 unpruned tree

------------------

age = 20-24

| Ln_inc <= 7.467942: Students (98.0/61.0)

| Ln_inc > 7.467942: Employed (218.0/83.0)

age = 15-19: Students (258.0/82.0)

Number of Leaves : 3

Size of the tree: 5

Спецификация дерева LMT

Logistic model tree

------------------

age = 20-24

| Ln_inc <= 7.467942: LM_1:1/3 (98)

| Ln_inc > 7.467942: LM_2:1/3 (218)

age = 15-19: LM_3:1/2 (258)

Number of Leaves : 3

Size of the Tree: 5

Доступность добровольного Медицинского страхования в России

Таблица 11

Спецификация регрессионной модели

	Insurance paid by employer	Insurance self-paid
City_sizeCity or Town	-0.214	-0.295***
City_sizeSmall_town_or_village	-1.099***	-1.534***
MaritalDivorced or Widowed	-0.584**	0.441***
MaritalWas never married	-0.346***	0.182***
Male	-0.027	-0.388***
Is_russian	-0.326	-1.064***
v_age	0.017	0.210***
age_sq	-0.0004	-0.002***
Ln_inc	0.871***	-0.049
educHigher	0.503**	0.238***
educProfessional	-0.094	-0.602***
HealthBad	-0.230***	-0.858***
HealthGood	-0.182	0.147***
Children_under_18	0.204*	0.230
professionOfficials or Directors	1.704***	2.404***
professionQualified workers	1.924***	1.021***
professionSpecialist w/ higher ed	1.689***	2.188***
professionSpecialists w/ sec ed	1.738***	1.311***
professionUnqualified workers	1.485***	0.923***
Constant	-12.161***	-10.854***
AIC	2,208.950	2,208.950
*p <.05; **p <.01; ***p <.001

Спецификация дерева Rpart

n= 14196

node), split, n, loss, yval, (yprob)

* denotes terminal node

1) root 14196 2839.200000 Insurance self-paid (0.110000000 0.090000000 0.800000000)

2) v_age>=19.5 10727 2434.832000 Insurance self-paid (0.132916391 0.143046788 0.724036821)

4) v_age>=73.5 1005 11.650210 No insurance (0.906804564 0.093195436 0.000000000) *

5) v_age< 73.5 9722 2309.824000 Insurance self-paid (0.121794034 0.143763254 0.734442712)

10) profession=Qualified workers 1628 523.163900 Insurance self-paid (0.164524134 0.331124380 0.504351486)

20) v_age< 52.5 1307 139.604800 Insurance paid by employer (0.329962199 0.670037801 0.000000000)

40) Ln_inc< 8.800001 493 42.717450 No insurance (0.560732660 0.439267340 0.000000000) *

41) Ln_inc>=8.800001 814 85.075210 Insurance paid by employer (0.261090019 0.738909981 0.000000000) *

21) v_age>=52.5 321 100.070600 Insurance self-paid (0.053845038 0.104389190 0.841765772) *

11) profession=Unemployed,Officials or Directors,Specialist w/ higher ed,Specialists w/ sec ed,Unqualified workers 8094 1786.660000 Insurance self-paid (0.115892541 0.117886640 0.766220819)

22) City_size=Small_town_or_village 2564 380.217400 Insurance self-paid (0.314514522 0.102131308 0.583354170)

44) v_age>=54.5 872 0.000000 No insurance (1.000000000 0.000000000 0.000000000) *

45) v_age< 54.5 1692 281.566400 Insurance self-paid (0.231429995 0.114510171 0.654059833)

90) v_age< 34.5 671 42.717450 No insurance (0.636089604 0.363910396 0.000000000)

180) Ln_inc< 8.406469 418 7.766809 No insurance (0.858346665 0.141653335 0.000000000) *

181) Ln_inc>=8.406469 253 27.604190 Insurance paid by employer (0.441279929 0.558720071 0.000000000) *

91) v_age>=34.5 1021 164.181800 Insurance self-paid (0.163233869 0.072479465 0.764286666)

182) Ln_inc>=7.829703 711 129.337200 Insurance self-paid (0.257028166 0.164557888 0.578413945)

364) v_age< 46.5 445 38.834040 No insurance (0.558936779 0.441063221 0.000000000) *

365) v_age>=46.5 266 41.290780 Insurance self-paid (0.135505439 0.053260361 0.811234200) *

183) Ln_inc< 7.829703 310 34.844630 Insurance self-paid (0.089403755 0.000000000 0.910596245) *

23) City_size=Moskow_or_spb,City or Town 5530 1406.442000 Insurance self-paid (0.088959777 0.120023034 0.791017189)

46) v_age< 22.5 274 15.533620 No insurance (0.662893358 0.337106642 0.000000000) *

47) v_age>=22.5 5256 1360.363000 Insurance self-paid (0.085003020 0.118526437 0.796470543)

94) Marital=Married 3636 1005.127000 Insurance self-paid (0.097357351 0.152563854 0.750078795)

188) v_age< 27.5 283 30.658750 Insurance paid by employer (0.396827722 0.603172278 0.000000000) *

189) v_age>=27.5 3353 927.867900 Insurance self-paid (0.091491752 0.143737978 0.764770271)

378) Health=Bad 332 19.417020 No insurance (0.655794692 0.344205308 0.000000000) *

379) Health=Average,Good 3021 871.456700 Insurance self-paid (0.083304480 0.140829468 0.775866051)

758) profession=Unemployed,Specialists w/ sec ed 2103 516.237800 Insurance self-paid (0.144765904 0.181773068 0.673461028)

1516) v_age>=56.5 598 31.067230 No insurance (0.682388066 0.317611934 0.000000000) *

1517) v_age< 56.5 1505 418.422700 Insurance self-paid (0.109308589 0.172814208 0.717877203)

3034) Ln_inc>=8.93378 517 231.695500 Insurance self-paid (0.090452240 0.304531155 0.605016605)

6068) v_age>=39.5 263 27.038530 Insurance paid by employer (0.224870911 0.775129089 0.000000000) *

6069) v_age< 39.5 254 111.455200 Insurance self-paid (0.055795114 0.183197028 0.761007858) *

3035) Ln_inc< 8.93378 988 186.727300 Insurance self-paid (0.121646249 0.086632151 0.791721601)

6070) v_age< 34.5 355 19.417020 No insurance (0.670971535 0.329028465 0.000000000) *

6071) v_age>=34.5 633 127.714100 Insurance self-paid (0.082939507 0.069552341 0.847508152) *

759) profession=Officials or Directors,Specialist w/ higher ed,Unqualified workers 918 355.218900 Insurance self-paid (0.041189364 0.112773748 0.846036888) *

95) Marital=Divorced or Widowed,Was never married 1620 355.235500 Insurance self-paid (0.066338567 0.067104003 0.866557430)

190) Ln_inc< 9.716201 1201 218.364900 Insurance self-paid (0.143218101 0.092047106 0.764734794)

380) Health=Average 650 34.950640 No insurance (0.674781591 0.325218409 0.000000000)

760) profession=Unemployed,Officials or Directors 369 0.000000 No insurance (1.000000000 0.000000000 0.000000000) *

761) profession=Specialist w/ higher ed,Specialists w/ sec ed,Unqualified workers 281 30.771880 Insurance paid by employer (0.468209118 0.531790882 0.000000000) *

381) Health=Bad,Good 551 110.896700 Insurance self-paid (0.073611176 0.061513900 0.864874924) *

191) Ln_inc>=9.716201 419 136.870600 Insurance self-paid (0.025185077 0.053752000 0.921062923) *

3) v_age< 19.5 3469 404.368100 Insurance self-paid (0.072368648 0.002891068 0.924740283)

6) City_size=Moskow_or_spb,Small_town_or_village 1453 171.355600 Insurance self-paid (0.154532312 0.007336829 0.838130858)

12) v_age>=10.5 602 3.883404 No insurance (0.945970549 0.054029451 0.000000000) *

13) v_age< 10.5 851 99.479880 Insurance self-paid (0.096882116 0.003935631 0.899182253) *

7) City_size=City or Town 2016 233.012500 Insurance self-paid (0.052208332 0.001800222 0.945991446) *

Спецификация дерева CHAID

Model formula:

TARGET ~ City_size + Marital + Male + Is_russian + educ + Health +

profession

Fitted party:

[1] root

| [2] profession in Unemployed

| | [3] educ in Secondary or lower, Professional: No insurance (n = 4379, err = 0.3%)

| | [4] educ in Higher

| | | [5] Marital in Married

| | | | [6] City_size in Moskow_or_spb: No insurance (n = 112, err = 3.6%)

| | | | [7] City_size in City or Town, Small_town_or_village: No insurance (n = 392, err = 0.5%)

| | | [8] Marital in Divorced or Widowed, Was never married: No insurance (n = 370, err = 0.8%)

| [9] profession in Officials or Directors, Qualified workers, Specialist w/ higher ed, Specialists w/ sec ed

| | [10] City_size in Moskow_or_spb

| | | [11] educ in Secondary or lower, Professional: No insurance (n = 328, err = 5.8%)

| | | [12] educ in Higher: No insurance (n = 292, err = 15.1%)

| | [13] City_size in City or Town

| | | [14] educ in Secondary or lower, Professional

| | | | [15] Male in 0: No insurance (n = 1018, err = 3.5%)

| | | | [16] Male in 1: No insurance (n = 1164, err = 6.4%)

| | | [17] educ in Higher

| | | | [18] Is_russian in 0: No insurance (n = 141, err = 12.1%)

| | | | [19] Is_russian in 1: No insurance (n = 1335, err = 7.8%)

| | [20] City_size in Small_town_or_village: No insurance (n = 1490, err = 2.2%)

| [21] profession in Unqualified workers

| | [22] Is_russian in 0: No insurance (n = 50, err = 6.0%)

| | [23] Is_russian in 1: No insurance (n = 358, err = 1.4%)

Number of inner nodes: 10

Number of terminal nodes: 13

Спецификация дерева C50

Call:

C5.0.formula(formula = TARGET ~., data = train, control = C5.0Control(earlyStopping = T,

CF = 0.53, fuzzyThreshold = F, noGlobalPruning = T))

Class specified by attribute `outcome'

Read 14196 cases (12 attributes) from undefined.data

Decision tree:

profession = Unemployed: No insurance (6530.7/36.6)

profession in {Officials or Directors,Qualified workers,

: Specialist w/ higher ed,Specialists w/ sec ed,

: Unqualified workers}:

:...v_age <= 19: No insurance (1424.9/14.6)

v_age > 19:

:...Ln_inc <= 8.944289: No insurance (3084/79.3)

Ln_inc > 8.944289:

:...Ln_inc <= 11.05618: No insurance (3112.9/242.4)

Ln_inc > 11.05618:

:...Is_russian <= 0: Insurance paid by employer (4.1/0.1)

Is_russian > 0:

:...Health in {Bad,Good}: No insurance (18.6/5.1)

Health = Average: [S1]

Спецификация дерева C4.5/J48

J48 unpruned tree

------------------

Ln_inc <= 8.944289

| profession = Unemployed

| | Is_russian <= 0: No insurance (398.0/2.0)

| | Is_russian > 0

| | | Children_under_18 <= 0

| | | | Health = Average

| | | | | educ = Secondary or lower

| | | | | | City_size = Moskow_or_spb: No insurance (38.0)

| | | | | | City_size = City or Town: No insurance (225.0/1.0)

| | | | | | City_size = Small_town_or_village: No insurance (230.0/1.0)

| | | | | educ = Higher: No insurance (145.0/1.0)

| | | | | educ = Professional: No insurance (238.0)

| | | | Health = Bad: No insurance (462.0)

| | | | Health = Good: No insurance (117.0)

| | | Children_under_18 > 0

| | | | Health = Average: No insurance (203.0/3.0)

| | | | Health = Bad: No insurance (31.0)

| | | | Health = Good: No insurance (245.0)

| profession = Officials or Directors: No insurance (100.0/2.0)

| profession = Qualified workers

| | City_size = Moskow_or_spb: No insurance (24.0)

| | City_size = City or Town: No insurance (347.0/15.0)

| | City_size = Small_town_or_village: No insurance (231.0/4.0)

| profession = Specialist w/ higher ed: No insurance (341.0/11.0)

| profession = Specialists w/ sec ed

| | City_size = Moskow_or_spb: No insurance (73.0/4.0)

| | City_size = City or Town

| | | educ = Secondary or lower: No insurance (250.0/8.0)

| | | educ = Higher: No insurance (227.0/9.0)

| | | educ = Professional: No insurance (228.0/2.0)

| | City_size = Small_town_or_village

| | | v_age <= 40: No insurance (192.0/1.0)

| | | v_age > 40: No insurance (207.0)

| profession = Unqualified workers: No insurance (206.0/2.0)

Ln_inc > 8.944289

| v_age <= 59

| | profession = Unemployed: No insurance (173.0)

| | profession = Officials or Directors: No insurance (191.0/27.0)

| | profession = Qualified workers

| | | Health = Average: No insurance (279.0/28.0)

| | | Health = Bad: No insurance (19.0/1.0)

| | | Health = Good: No insurance (296.0/19.0)

| | profession = Specialist w/ higher ed: No insurance (369.0/43.0)

| | profession = Specialists w/ sec ed

| | | Health = Average

| | | | Ln_inc <= 9.36743: No insurance (208.0/12.0)

| | | | Ln_inc > 9.36743: No insurance (195.0/27.0)

| | | Health = Bad: No insurance (20.0)

| | | Health = Good: No insurance (320.0/30.0)

| | profession = Unqualified workers: No insurance (78.0/4.0)

| v_age > 59

| | Health = Average

| | | educ = Secondary or lower: No insurance (269.0/1.0)

| | | educ = Higher: No insurance (252.0/8.0)

| | | educ = Professional: No insurance (231.0/3.0)

| | Health = Bad: No insurance (437.0)

| | Health = Good: No insurance (98.0/4.0)

Number of Leaves : 39

Size of the tree: 58

Спецификация дерева LMT

Logistic model tree

------------------

: LM_1:14/14 (8193)

Number of Leaves : 1

Size of the Tree: 1

LM_1:

Class No insurance:

3.1 +

[City_size=Moskow_or_spb] * -0.15 +

[City_size=Small_town_or_village] * 0.46 +

[Is_russian] * 0.2 +

[age_sq] * 0 +

[Ln_inc] * -0.11 +

[educ=Higher] * -0.43 +

[educ=Professional] * 0.08 +

[Health=Average] * -0.08 +

[Children_under_18] * -0.15 +

[profession=Unemployed] * 1.07 +

[profession=Unqualified workers] * 0.15

Class Insurance paid by employer:

-6.92 +

[City_size=Small_town_or_village] * -0.49 +

[Marital=Married] * 0.25 +

[Marital=Divorced or Widowed] * -0.15 +

[age_sq] * -0 +

[Ln_inc] * 0.69 +

[Children_under_18] * 0.06 +

[profession=Unemployed] * -0.69 +

[profession=Qualified workers] * 0.1

Class Insurance self-paid:

-3.32 +

[City_size=Small_town_or_village] * -0.25 +

[Marital=Divorced or Widowed] * 0.23 +

[Male] * -0.26 +

[Is_russian] * -0.66 +

[v_age] * 0.02 +

[educ=Higher] * 0.18 +

[Health=Bad] * -0.35 +

[profession=Officials or Directors] * 0.72 +

[profession=Specialist w/ higher ed] * 0.5

Внутрифирменная мобильность на российском рынке труда

Таблица 11

Спецификация регрессионной модели

	External	Horizontal	Mixed	Vertical
	-1	-2	-3	-4
Male	0.208*	0.188	0.516	0.185
Agegroup30-39	-0.388***	0.039	-0.139	-0.731***
Agegroup40-49	-0.642***	-0.059	-0.728	-1.022***
Agegroup50 plus	-1.051***	-0.545	-1.134	-1.623***
educationHigher	-0.226***	-0.498*	-0.926	-0.325*
educationLess than secondary	-0.354**	-0.849*	-10.922	-0.896**
educationProfessional	-0.294***	-0.483**	-0.718	-0.431**
Married	-0.258**	-0.127	0.298	-0.071
City_sizeCity	0.317*	-0.300	-0.703	-0.153
City_sizeMoskow_or_spb	0.323	-0.081	0.001	-0.439
City_sizeTown	0.046	-0.452	-0.434	-0.283
Firm_size101-500	-0.278	1.121**	1.726**	0.090
Firm_size501-100	-0.448	0.658	1.111	-0.536
Firm_size51-100	-0.415*	0.464	1.269	0.029
Firm_sizeMore than 1000	-0.106	0.603	1.379*	0.295
OwnershipForeign	0.544	-7.597	0.063	-0.396
OwnershipPrivate	0.747***	-0.227	-0.953*	-0.081
Professional_statusBlue_collar_qual	-0.446*	-0.422	-0.266	0.210
Professional_statusWhite_collar_qual	-0.454*	-0.111	1.209	0.937
Professional_statusWhite_collar_unqual	-0.254	0.247	0.149	1.032
sectorCommercial_services	0.343	0.114	9.072	0.254
sectorConstruction	0.644*	0.136	8.502	-0.314
sectorIndustry	0.157	0.837	9.380	0.076
sectorNon-comercial_services	0.134	0.003	8.422	0.098
Subordinates1-49	-0.458**	-0.248	0.321	1.013***
Subordinates50 plus	-0.168	0.468	0.955	1.317**
educFull secondary	-0.440***	-1.422***	-0.812	-1.060***
educHigher	-0.226***	-0.498*	-0.926	-0.325*
educProfessional	-0.294***	-0.483**	-0.718	-0.431**
Constant	-1.314***	-3.251***	-13.378	-2.710***
AIC	6,126.268	6,126.268	6,126.268	6,126.268
*p <.05; **p <.01; ***p <.001

Спецификация дерева Rpart

n= 6024

node), split, n, loss, yval, (yprob)

* denotes terminal node

1) root 6024 3012.00000 Immobile (0.5 0.1 0.15 0.1 0.15)

2) Agegroup=50 plus 1510 420.52830 Immobile (0.66 0.071 0.12 0.062 0.08) *

3) Agegroup=Less than 30,30-39,40-49 4514 2591.47200 Immobile (0.46 0.11 0.16 0.11 0.17)

6) Subordinates=No subordinates 3613 1908.50700 Immobile (0.48 0.12 0.18 0.1 0.12)

12) Firm_size=Less than 50 1438 591.07070 Immobile (0.54 0.15 0.11 0.048 0.15) *

13) Firm_size=101-500,501-100,51-100,More than 1000 2175 1317.43700 Immobile (0.45 0.11 0.22 0.13 0.1)

26) sector=Commercial_services,Construction,Non-comercial_services 1517 840.02080 Immobile (0.47 0.12 0.16 0.13 0.12)

52) City_size=City,Town 1058 484.56930 Immobile (0.52 0.13 0.15 0.062 0.14) *

53) City_size=Small_town_or_village,Moskow_or_spb 459 355.45140 Immobile (0.38 0.099 0.18 0.24 0.098)

106) Profe...