The f-wald test, Advanced Statistics

Primary Model

Below is a regression analysis without 17 outliers that have been removed

Regression Analysis: wfood versus totexp, income, age, nk

The regression equation is

wfood = 0.378 - 0.00129 totexp - 0.000054 income + 0.00170 age + 0.0317 nk

Predictor              Coef       SE Coef           T          P         VIF

Constant         0.37816     0.01356         27.89  0.000

totexp         -0.00128554  0.00006284  -20.46  0.000    1.324

income        -0.00005410  0.00004950   -1.09   0.275    1.341

age              0.0016993    0.0003058      5.56   0.000    1.065

nk                0.031717      0.004676        6.78   0.000    1.007

S = 0.0880161   R-Sq = 28.0%   R-Sq(adj) = 27.8%

Analysis of Variance

Source               DF       SS          MS         F           P

Regression         4      4.5159    1.1290  145.73  0.000

Residual Error   1497  11.5970   0.0077

  Lack of Fit       1328   10.1731  0.0077    0.91  0.806

  Pure Error        169    1.4239    0.0084

Total                  1501  16.1129


Secondary Model

Below is a regression analysis without 17 outliers that have been removed and dropping the income variable   

Regression Analysis: wfood versus totexp, age, nk

The regression equation is

wfood = 0.376 - 0.00132 totexp + 0.00165 age + 0.0317 nk

Predictor         Coef     SE Coef       T      P    VIF

Constant       0.37593     0.01341   28.04  0.000

totexp     -0.00131710  0.00005581  -23.60  0.000  1.045

age          0.0016462   0.0003019    5.45  0.000  1.038

nk            0.031672    0.004676    6.77  0.000  1.007

S = 0.0880218   R-Sq = 28.0%   R-Sq(adj) = 27.8%

Analysis of Variance

Source               DF       SS          MS          F           P

Regression         3       4.5067   1.5022   193.89  0.000

Residual Error   1498  11.6063  0.0077

  Lack of Fit       644    4.9570    0.0077    0.99     0.560

  Pure Error      854     6.6493    0.0078

Total               1501    16.1129

The Null Hypothesis - H0: No difference between the primary and secondary model

1465_The F-Wald Test.png

Since the F value is 1.2005 < 3.8477 there is sufficient evidence to suggest that we accept H0 implying that there is no difference between the primary and secondary model and income can be removed.

Posted Date: 3/4/2013 7:10:51 AM | Location : United States

Related Discussions:- The f-wald test, Assignment Help, Ask Question on The f-wald test, Get Answer, Expert's Help, The f-wald test Discussions

Write discussion on The f-wald test
Your posts are moderated
Related Questions
Multi-hit model is the model for a toxic response which results from the random occurrence of one or the more fundamental biological events. A response is supposed to be induced o

The procedures used for determining how the quality of life is affected by the environment, in particular by factors such as air and solid wastes, water pollution, hazardous substa

a. Explain the meaning of the word non-orthogonal. b. What condition(s) must exist for non-orthogonality to occur? Be specific.

O'Brien's two-sample tests are the extensions of the conventional tests for assessing the differences between treatment groups which take account of the possible heterogeneous nat

Ordinal variable is a measurement which allows a sample of the individuals to be ranked with respect to some characteristic but where differences at different points of the scale

Discuss the use of dummy variables in both multiple linear regression and non-linear regression. Give examples if possible

The particular projection which an investigator believes is most likely to give an accurate prediction of the future value of some process. Commonly used in the context of the anal

The nonparametric Bayesian inference approach to using the finite mixture distributions for modelling data suspected of the containing distinct groups of observations; this approac

Last observation carried forward is a technique for replacing the observations of the patients who drop out of the clinical trial carried out over a time period. It consists of su

Range is the difference between the largest and smallest observations in the data set. Commonly used as an easy-to-calculate measure of the dispersion in the set of observations b