Lagrange multipliertest, Advanced Statistics

The Null Hypothesis - H0:  There is autocorrelation

The Alternative Hypothesis - H1: There is no autocorrelation

Rejection Criteria: Reject H0 (n-s)R2 >641_Partial Autocorrelation Function1.png = (1515 - 4) x (0.01) = 15.11 > 9.49 (641_Partial Autocorrelation Function1.png)

1515 cases used, 4 cases contain missing values

Since 15.11 > 9.49 the chi-squared value with 4 lags (ET-1, ET-2, ET-3, and ET-4) there is evidence to suggest that we reject H0 meaning that there is no autocorrelation.    

The regression equation is

RESI1 = - 0.0011 + 0.000005 totexp - 0.000001 income + 0.000017 age + 0.00007 nk

        + 0.0085 ET-1 + 0.0070 ET-2 - 0.0284 ET-3 - 0.0074 ET-4

Predictor         Coef     SE Coef                 T      P

Constant       -0.00105     0.01375        -0.08  0.939

totexp          0.00000471  0.00006080   0.08  0.938

income        -0.00000082  0.00004314  -0.02  0.985

age              0.0000167   0.0003090     0.05  0.957

nk                0.000071     0.004785       0.01  0.988

ET-1             0.00847       0.02580         0.33  0.743

ET-2             0.00700       0.02584         0.27  0.786

ET-3           -0.02842       0.02587        -1.10  0.272

ET-4          -0.00743       0.02592         -0.29  0.774

As the T value decreases, the P value increases which is noticeable above due to the inclusions of lags. Most of the T values are now closer to 0 which shows that there is less reliability of the coefficient.  ET-3 will be included in a further regression analysis as it is significant with a value of -1.10, conversely ET-1, ET-2, ET-4 will be removed as they are insignificant with low T values.     

S = 0.0905514   R-Sq = 0.1%   R-Sq(adj) = 0.0%

The inclusion of lags has caused the r-squared to be really low at 0.1% which certainly suggests that the model is inadequate for explaining the Y variable. It also indicates that data points are distributed away from the line of best fit and that the independent variables are poor predictors for the dependent variable. The remaining percentage (99.9%) is the variation which is unknown.


Analysis of Variance


Source               DF         SS        MS     F      P

Regression        8    0.012127  0.001516  0.18  0.993

Residual Error  1506  12.348529  0.008200

Total                1514  12.360656

Source  DF    Seq SS

totexp   1  0.000029

income  1  0.000005

age       1  0.000011

nk         1  0.000000

ET-1      1  0.000903

ET-2      1  0.000544

ET-3      1  0.009961

ET-4      1  0.000673

Since the F value is small at 0.18 and the P value is high 0.993 it reveals that there is no relationship between the Y dependent variable and X independent variables. This indicates that as it is 0.18 it does not support the model and therefore the slopes are equal to 0.

Posted Date: 3/4/2013 6:39:55 AM | Location : United States

Related Discussions:- Lagrange multipliertest, Assignment Help, Ask Question on Lagrange multipliertest, Get Answer, Expert's Help, Lagrange multipliertest Discussions

Write discussion on Lagrange multipliertest
Your posts are moderated
Related Questions
Confidence interval : A range of the values, calculated from the sample observations which is believed, with the particular probability, to posses the true parameter value. A 95% c

Reinterviewing  is the second interview for a sample of survey respondents in which questions of the original interview (or the subset of them) are repeated again. The same methods

Multiple correlation coefficient is the correlation among the observed values of dependent variable in the multiple regression, and the values predicted by estimated regression

Write a c++ program to find the sum of 0.123 ? 10 3 and 0.456 ? 10 2 and write the result inthree significant digits.

Hot deck is a method broadly used in surveys for imputing the missing values. In its easiest form the method includes sampling with replacement m values from the sample respondent

Your first task is to realize two additional data generation functions. Firstly, extend the system to generate random integral numbers based on normal distribution. You need to stu

Harris and Stevens forecasting is the method of making short term forecasts in the time series which is subject to abrupt changes in pattern and the transient effects. Instances o

Why Graph theory? It is the branch of mathematics concerned with the properties of sets of points (vertices or nodes) some of which are connected by the lines known as the edges. A

VIF is the abbreviation of variance inflation factor which is a measure of the amount of multicollinearity that exists in a set of multiple regression variables. *The VIF value