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
A law supposedly applicable to voting behaviour which has a history of several decades. It may be stated thus: Consider a two-party system and suppose that the representatives of t

The scatter plot of SRES1 versus totexp demonstrates that there is non-linear relationship that exists as most of the points are below and above zero. The scatter plot show that th

Greenhouse geissercorrection is the method of adjusting the degrees of freedom of the within- subject F-tests in the analysis of the variance of longitudinal data so as to allow t

need answers to questions in book advanced and multivariate statistical methods

Human capital model : The model for evaluating the economic implication of the disease in terms of the economic loss of a person succumbing to morbidity or the mortality at some pa

Mardia's multivariate normality test is a test that a set of the multivariate data arise from the multivariate normal distribution against departures due to the kurtosis. The test

Jonckheere Terpstra test  is the test for detecting particular types of departures from the independence in a contingency table in which both the row and column categories contain

Interior analysis is the  term now and again applied to analysis carried out on the fitted model in regression problem. The basic target of such analyses is the identification of

Hosmer-Lemeshow test is a goodness-of-fit test taken in use in logistic regression, particularly when there are regular covariates. Units are spitted into deciles based on predict

Pie chart is an extensively used graphical technique for presenting relative frequencies related with the observed values of the categorical variable. The chart comprises of a cir