Use a non-linear estimators such as Probit or Logit (or both) to estimate a baseline model in which the

INSTRUCTIONS TO CANDIDATES

ANSWER ALL QUESTIONS

Task 1: Is there evidence of discrimination in mortgage lending? (50%).

You are given a dataset called Wldrg_Mortgage_App, with 1989 observations on approvals and rejections of mortgage applications in a US urban area. Description of all variables is given below.

1.  occ       occupancy 2.  loanamt   loan amt in thousands 3.  action    type of action taken 4.  msa       msa number of property 5.  suffolk   =1 if property in Suffolk County 6.  race      race of applicant 7.  gender    gender of applicant 8.  appinc    applicant income, $1000s 9.  typur     type of purchaser of loan 10. unit      number of units in property 11. married   =1 if applicant married 12. dep       number of dependents 13. emp       years employed in line of work 14. yjob      years at this job 15. self      self-employment dummy 16. atotinc   total monthly income 17. cototinc coapp total monthly income 18. hexp      propose housing expense 19. price     purchase price 20. other     other financing, $1000s 21. liq       liquid assets 22. rep       no. of credit reports 23. gdlin     credit history meets guidelines 24. lines     no. of credit lines on reports 25. mortg     credit history on mortgage paym 26. cons      credit history on consumer stuf 27. pubrec    =1 if filed bankruptcy 28. hrat      housing exp, % total inccome 29. obrat     other oblgs, % total income 30. fixadj    fixed or adjustable rate? 31. term      term of loan in months

32. apr       appraised value 33. prop      type of property 34. inss      PMI sought 35. inson     PMI approved 36. gift      gift as down payment 37. cosign    is there a cosigner 38. unver     unverifiable info 39. review    number of times reviewed 40. netw      net worth 41. unem      unemployment rate by industry 42. min30     =1 if minority pop. > 30% 43. bd        =1 if boarded-up val > MSA med 44. mi        =1 if tract inc > MSA median 45. old       =1 if applic age > MSA median 46. vr        =1 if tract vac rte > MSA med 47. sch       =1 if > 12 years schooling 48. black     =1 if applicant black 49. hispan    =1 if applicant Hispanic 50. male      =1 if applicant male 51. reject    =1 if action == 3 52. approve   =1 if action == 1 or 2 53. mortno    no mortgage history 54. mortperf no late mort. payments 55. mortlat1 one or two late payments 56. mortlat2 > 2 late payments 57. chist     =0 if accnts deliq. >= 60 days 58. multi     =1 if two or more units 59. loanprc   amt/price 60. thick     =1 if rep > 2 61. white     =1 if applicant white 62. obwhte    obrat*awhite

Please carry out the following tasks.

1. Use a non-linear estimators such as Probit or Logit (or both) to estimate a baseline model in which the probability of rejection depends only on racial and demographic characteristics of the applicant. Carry out post-estimation evaluation and interpret your findings. (10 marks)

2. Add to model (1) above a range of credit risk indicators associated with the applicant, re- estimate and run post-estimation tests as above. Interpret your findings and compare with those in (1) above. (15 marks).

3. Add to model (2) above a range of credit risk indicators associated with the property, re- estimate and run post-estimation tests as above. Interpret your findings and compare with those in (1) and (2) above. (15 marks).

4. Conclude by identifying the ‘best’ model and interpreting your findings in the light of relevant literature (10 marks).

Relevant works to start with and relevant Stata routines:

Hunter, W. C., & Walker, M. B. (1996). The cultural affinity hypothesis and mortgage lending decisions. The Journal of Real Estate Finance and Economics, 13(1), 57-70. Available here.

Robinson, J. K. (2002). Race, gender, and familial status: discrimination in one US mortgage lending market. Feminist Economics, 8(2), 63-85. Available here .

Kau, J. B., Keenan, D. C., & Munneke, H. J. (2012). Racial discrimination and mortgage lending. The Journal of Real Estate Finance and Economics, 45(2), 289-304. Available here .

Probit/Logit model specification test (linktest). See Stata documentation available here.

Model performance: area under receivers operating characteristics (ROC) curve. See Stata documentation available here .

Task 2: Testing for weak-form market efficiency and volatility (50%)

You are given the dataset Wldrg_NYSE, which contains 691 weekly time-series observations on New York Stock Exchange (NYSE) stock price and returns.

Obs: 691

1. price NYSE stock price index

2. return 100*(p - p(-1])/p(-1))

3. return_1 lagged return

4. t time trend: 1 to 691

5. price_1 price(-1)

6. price_2 price(-2)

7. cprice price - price_1

8. cprice_1 cprice(-1)

Drawing on the relevant theoretical and empirical literature:

1. Test for weak-form market efficiency, using four methods: serial correlation, runs, variance ratio, and unit root. (15 marks)

2. Comment on strengths and weaknesses of the efficient market hypothesis and the method used (10 marks)

3. Estimate return volatility using different ARCH and GARCH models. (15 marks)

4. Conclude by interpreting your findings and commenting on the strengths and weaknesses of the theory and the method used (10 marks)

Two relevant work to kick off with:

Degutis, A., & Novickyte, L. (2014). The efficient market hypothesis: A critical review of literature and methodology. Ekonomika, 93(2), 7. Available here

Pilbeam, K., & Langeland, K. N. (2015). Forecasting exchange rate volatility: GARCH models versus implied volatility forecasts. International Economics and Economic Policy, 12(1), 127-142. Available here

NOTE on producing regression output tables Use esttab to produce output tables as word files. This is how:

Estimate model1

Type: estimates store model1 (or any meaningful name you choose) Estimate model2

Type: estimates store model2 (or any meaningful name you choose) Finally:

esttab model1 model 2 using Drive:\folder\file_name.rtf, ///

cells(b(star fmt(3)) se(par fmt(3))) starlevels(* 0.10 ** 0.05 *** 0.01) /// stats (N k df_m ll aic bic) varwidth(15) modelwidth(10)

Notes:

• The /// are used to break long lines in the DO file. There must be a space before and after each ///

• The width can be adjusted if necessary

• The fmt(3) specifies the decimal place. 3 decimal places are OK. The output will be stored as rtf file in the directory you specify.

Attachments:

Related Questions

. The fundamental operations of create, read, update, and delete (CRUD) in either Python or Java

CS 340 Milestone One Guidelines and Rubric  Overview: For this assignment, you will implement the fundamental operations of create, read, update,

. Develop a program to emulate a purchase transaction at a retail store. This  program will have two classes, a LineItem class and a Transaction class

Retail Transaction Programming Project  Project Requirements:  Develop a program to emulate a purchase transaction at a retail store. This

. The following program contains five errors. Identify the errors and fix them

7COM1028   Secure Systems Programming   Referral Coursework: Secure

. Accepts the following from a user: Item Name Item Quantity Item Price Allows the user to create a file to store the sales receipt contents

Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip

. The final project will encompass developing a web service using a software stack and implementing an industry-standard interface. Regardless of whether you choose to pursue application development goals as a pure developer or as a software engineer

CS 340 Final Project Guidelines and Rubric  Overview The final project will encompass developing a web service using a software stack and impleme