Pooja G. Mookim





Welcome to my homepage!!

I am a doctoral candidate at Boston University.




Boston University
Department of Economics
270 Bay State Road

Boston MA 02215




Cell:   (857) 472-0235

Email: pgupta1@bu.edu



Curriculum Vitae (HTML)

Curriculum Vitae (PDF)


The woods are lovely, dark and deep,
But I have promises to keep,
And miles to go before I sleep,
And miles to go before I sleep.

                                              - Robert Frost

Fields of Interest: Industrial Organization, Health Economics, Econometrics.




Prof. Randall Ellis                                      

Department of Economics   

Boston University                   

270 Bay State Road

Boston MA 02215


Ph: (617) 353-2741

Fax: (617) 353-4449

Prof. Kevin Lang

Department of Economics

Boston University

270 Bay State Road

Boston MA 02215


Ph: (617) 353-5694

Fax: (617) 353-4449

Prof. Maristella Botticini       

Department of Institutional Analysis and Public Management       

Bocconi University                                

Via Sarfati                                          
Milano- PI 03628350153                                        maristella.botticini@unibocconi.it                                  




Infertility Treatment, ART and IUI Procedures and Delivery Outcomes: How Important is Selection? (with Randall Ellis)

 Female fertility rates are viewed differently in developed and developing countries. While high fertility levels are a cause of concern in developing countries like India, women in the United States worry more about undesirable infertility. In this paper, I use medical claims data to examine maternal and neonatal outcomes following the use of Assisted Reproductive Technologies (ARTs). Using two-stage analysis I show that once the endogeneity problem in ART is corrected using state mandates as instruments, health outcomes such as complications during pregnancy, miscarriage, abortion and ectopic pregnancies, and poor health of babies are only weakly associated with ART but are more significantly correlated with indicators of the mother’s age and health status before conception.


Gender Preferences, Fertility Choices and Government Policy in India


This paper evaluates the impact of the Indian government’s 1997 target-free policy on Indian’s high fertility rates. Using household longitudinal data spanning 30 years, I model the likelihood that a woman will give birth given her past birth history. The new policy reduces the likelihood of third and higher order births, suggesting that women are more likely to have smaller families after the policy change. However, women are more likely to have first and second births and are more likely to have them sooner in life. The policy has not been effective in reducing the “son preference” bias as mothers without sons remain more likely to have another child sooner than mothers with sons


 Cross-Validation Methods for Risk Adjustment Models” (with Randall Ellis)


This paper takes a fresh look at cross-validation techniques for assessing the predictive validity of risk adjustment models within the classical linear framework. I show that a K-Fold cross-validation is more efficient than 50-50 split sample technique and illustrate that biases in measures of goodness of fit in rich risk adjustment models remain meaningful in samples of up to 500,000 observations. A new estimation algorithm is described that calculates K-Fold cross-validated R-squared efficiently, so that it can be applied easily on sample sizes in the millions without resorting to split-sample techniques. The density functions obtained in repeated samples using this technique are statistically similar to those using conventional split sample methods.


Works in progress:

Health Premium Payment Systems in Private establishments”


There are many ways in which employers can share the cost of health insurance premiums with their employees. This paper develops a model of health premium cost sharing to predict the relationship between premium cost sharing and premium levels. Cross-section data from several large and small firms is used to quantify the relationship between single and family premium cost sharing parameters and health plan premiums.