Question

A random sample of Denver neighborhoods was obtained. The data (house) are for each neighborhood POP total population (in thousands) POPCH percentage change in population over past several years CHILD percentage of children (under 18) in population LUNCH percentage free school lunch participation INCOME percentage change in household income over past several years CRIME = crime rate (per 1000 population) CRIMECH = percentage change in crime rate over past several years Reference: The Piton Foundation, Denver, Colorado The following model was fit: Bo+B1Xi1 + B2X12 + B3X13 + B4Xi4 + B5X15 B6Xi6 +Ei Yi or Bo+B1POPB2POPCH B3CHILD + B4LUNCH + B5INCOME B6CRIMECHE CRIME

I have completed the table. Please provide interpretation for every number in the above table. Thank you!

Answer 1

DF- degrees of freedom

Using 6 predictors, we calculate one response CRIME. Each predictor is with 1 DF and residuals have 37 DF.

Then we calculate Sum of square for each predictor.

Mean squares are the estimate of population variance, (mean square is also known as sample variance). If we divide sum of squares by it's degrees of freedom then we get mean Sum of square

F value is the ratio of two mean square values. If the null hypothesis is true, we expect F value is almost 1.

From the probability of observing F statistic, we conclude about the null hypothesis whether we should accept or reject the given null hypothesis. If probability is low , then we can reject the null hypothesis for given population.

Significant codes at the bottom is that how many decimal value we have. It's a kind of notation for identication.