WWW.DISSERS.RU


...
    !

Pages:     | 1 |   ...   | 10 | 11 || 13 | 14 |   ...   | 39 |

0.2763

0.1212

0.1484

0.1569

0.0128

0.1756

Note. coefficientsestimating linear connection (association) of the ranges of each factor withFES and standard errors are applied here.

The lengthier chains of observations ofbarter demand have made it possible to assess how the FES has beeninfluenced by effective and barter demand since quarter IV of the year 1998.The fitting quality of the model utilizing linear interactions of the effectiveand barter demands with the actual FES estimates of the enterprises was lowerthan the fitting quality of the previous model but reliably higher than the 5%threshold. The model’s coefficients for the effective demand were positive and (exceptfor the two January surveys) statistically significant. The coefficients forthe barter demand were mostly negative and always statistically insignificant.And again the calculations confirm that the effective demand can be regarded asthe sole factor positively influencing an enterprise’s situation.

Now let us turn to the plans and forecastsof enterprises. In this case we are going to test the models where theformation of the expected changes in the FESs (again, the real, not thereported ones) occur under the influence of the forecasted values of changes indemand. The application of forecasted values, in our opinion, must moreaccurately reflect the enterprises’ own preferences. In this case the harsh market reality in a lessdefinite manner influences the connections under study.

At first we are going to assess the modelwhere only effective and barter demands are included. To achieve good fittingquality of the model with the participation of the expected FES changes, thecurrent FES estimated must be entered6, without utilizing anyinteractions including the participation of this variable. Themodel’s fittingquality turned out to be exceptionally high: the observed significance levelsin the majority of cases were close to 1 (see Table 17). The influence of thebarter demand forecasts was found to be stable, positive and statisticallysignificant during the whole period under analysis. The influence of theeffective demand forecasts was positive until the second quarter of 2001 butbecame significant only in late 1998 - early 1999. Nevertheless the influenceof the effective demand on the FES was twice as high. Beginning with the secondquarter of the past year, Russian industrial enterprises did not mention barteras a means to improve the values of their business activity even in theirforecasts.

Table 17

Characteristics of the influence offorecasted effective and barter demands on the expected changes of theenterprises’ FES

Date

Characteristics of model’s
fitting quality

Model’s coefficients



effective demand

barter demand


G2

df

Sig

SE

SE

10/98

152.1469

127

0.0636

0.6007

0.0867

0.2799

0.0850

1/99

119.2811

127

0.6745

0.7723

0.1118

0.3262

0.1147

4/99

116.4677

127

0.7382

0.8791

0.0995

0.0879

0.1047

7/99

87.7935

127

0.9968

1.1099

0.1187

0.1401

0.1001

10/99

86.0843

127

0.9979

0.9600

0.1049

0.0739

0.1164

1/00

90.4826

127

0.9940

1.0157

0.1207

0.2216

0.1239

4/00

130.2608

127

0.4034

1.1752

0.1122

0.0998

0.1122

7/00

88.8699

127

0.9959

1.1374

0.1340

0.0075

0.1242

10/00

82.7119

127

0.9992

0.8934

0.1115

0.0973

0.1042

1/01

88.8818

127

0.9959

1.0020

0.1197

0.1041

0.1170

4/01

54.2137

127

1.0000

1.2547

0.1217

-0.0341

0.1154

7/01

92.4163

127

0.9909

0.9664

0.1288

-0.0269

0.1187

10/01

64.1866

127

1.0000

0.9648

0.1466

-0.0085

0.1438

Note. This table presents: G2 -probability ratio; df - degrees of freedom; Sig - observed significancelevel; coefficients estimating the linear connection (association) of theranges of each factor with FES, and standard errors (SE).

Finally, we are going to consider themodels where all the types of the demand function as independent variables. Thefitting quality turns out to be very high (see Table 18). The model’s coefficients are positive andstatistically significant only for the effective demand. The forecasted valuesof other types of demand do not statistically significantly influence theexpected changes in the FES of Russian industrial enterprises. The coefficientsfor non-monetary types of demand were sometimes negative.

Table 18

Characteristics of the influence offorecasted effective,barter and other non-monetary types of demand on the expected changes of theenterprises’FES

Data

Characteristics of model’s fitting quality

Model’scoefficients



effective demand

barter demand

other non-monetary types of demand


G2

Df

Sig

SE

SE

SE

4/00

153.2533

396

1.0000

1.2206

0.1367

0.2838

0.1702

-0.0788

0.1912

7/00

107.3261

396

1.0000

1.2025

0.1648

-0.0570

0.2036

0.3144

0.2196

10/00

117.4724

396

1.0000

0.7667

0.1388

0.1697

0.1717

0.2761

0.1920

1/01

115.9058

396

1.0000

0.9010

0.1366

-0.1445

0.1743

0.4545

0.1939

4/01

76.8033

396

1.0000

1.2257

0.1411

0.0885

0.2027

-0.0040

Pages:     | 1 |   ...   | 10 | 11 || 13 | 14 |   ...   | 39 |



2011 www.dissers.ru -

, .
, , , , 1-2 .