Publications by Keller Whitney
PSet1
(2 p) Give a formal definition of a random variable (henceforth RV). An event that occurs with some probability. (4 p) Consider an experiment in which an Amazon.com customer is selected at random from the population of Amazon customers and her total spending during the month of December 2018 is measured. Suggest a discrete RV and a continuous R...
5672 sym
PSet#3
Premise 1.) Treated Sample Averages age: 25.8162162 edu: 10.3459459 black: 0.8432432 hisp: 0.0594595 married: 0.1891892 nodegree: 0.7081081 re74: 2095.5736934 re75: 1532.0553131 re78: 6349.1435021 u74: 0.7081081 u75: 0.6 Control Sample Averages age: 25.0538462 edu: 10.0884615 black: 0.8269231 hisp: 0.1076923 married: 0.1538462 ...
6269 sym R (13603 sym/35 pcs)
PSet#2
Problem One (50p) Let customeri’s spending at Walmart.com in a given month beyi. LetDi= 1 if customeriis aSam’s Club Plusmember and zero otherwise. Suppose thatyiis determined by a customer’sSam’s ClubPlus-membership status (Di) and other determinants (εi) according to the constant treatment effects model: \[yi=α+ρDi+εi\] For example...
11811 sym
Pset#4
Part One: A First Look at Observational Data 1.) CPS Comparison Averages age: 33.2252376 edu: 12.0275138 black: 0.0735368 hisp: 0.072036 married: 0.7117309 nodegree: 0.2958354 re74: 1.4016810^{4} re75: 1.365080410^{4} re78: 1.48466610^{4} u74: 0.1196223 u75: 0.1093047 2.) PSID Comparison Averages age: 34.8506024 edu: 12.1168675 ...
2990 sym R (3045 sym/12 pcs)
PSet#5
Part One 1.) a.) Estimating p-score using LPM LPM <- lm(treat ~ age + agesq + edu + edusq + married + nodegree + black + hisp + re74 + re75 + re74sq + u74black, data = data) data$p_hat_LPM <- predict(LPM) Estimating p-score using logit model logit <- glm(treat ~ age + agesq + edu + edusq + married + nodegree + black + hisp + re74 + re75 + re74s...
3110 sym R (3406 sym/8 pcs) 2 img
PSet#7
Part One: DD Estimator with Repeated Cross Sections 1.) I enjoyed reading and understanding slides 37 through 46 in the 4th set of slides! 2.) a.) Strating with equation (9) \[y_i = \theta_1(1-D_i)(1-T_i) + \theta_2D_i(1-T_i) + \theta_3(1-D_i)T_i + \theta_4D_iT_i + \mu_i\] Distribute \(\theta\)s and regroup in terms seen in equation (18) \[y_i...
7337 sym R (4269 sym/10 pcs)