To answer this question you need to use the data in APPLE.RAW. (i) Define a binary variable

Fantastic news! We've Found the answer you've been seeking!

Question:

To answer this question you need to use the data in APPLE.RAW.

(i) Define a binary variable ecobuy that takes the value 1 when ecolbs > 0 and 0 when ecobls = 0. In other words, at a given price, ecobuy marks whether a household buys environmentally friendly apples. What percentage of households claim to buy eco-apples?

(ii) Estimate the linear probability model

ecobuy =β_0+β_1 ecoprc+β_2 regprc+β_3 faminc+β_4 hhsize+β_5 educ+β_6 age+u

and report the results in the usual form. The coefficients of the price variables are carefully interpreted.

(iii) Are the non-price variables jointly significant in the LPM? (We use the usual F-statistic despite the fact that it is not valid in the presence of heteroskedasticity.) Which explanatory variable, other than the price variable, has the most important effect on the decision to purchase environmentally friendly apples? Do you think this is reasonable?

(iv) In the model in part (ii), replace faminc with log(faminc). which model fits the data better, using faminc or log(faminc)? Explain the coefficients of log(faminc).

(v) In the estimation in part (iv), how many of the estimated probabilities are negative? How many are greater than 1? Should this be brought to your attention?

(vi) For the estimates in part (iv), calculate the percentage of correct predictions for the outcomes ecobuy=0 and ecobuy=1. Which outcome is best predicted by the model?

(vii) Please regress the model in (iv) on the logit model, using only screenshots.

http://www.DeepL.com/Translator