1、University of Sydney Department of Economics ECMT2110 Semester 2 2012 Kadir Atalay ASSIGNMENT 2 DUE 4 pm Friday October 5 th Deadline 4pm Monday October 8 th INSTRUCTIONS Use STATA to conduct the following analysis Hand in your log file Use an editor to add your names and student numbers to the file
2、 and to add your written responses Iwould suggest after completing your log file you copy it into a Word document with nice font so the result come out try Courier New size 8 then type your answers after you have preformed the command HANDWRITTEN ASSIGNMENTS will get 0 point Q1 70 points Asset Prici
3、ng Models A well known model of the returns to financial assets is the Capital Asset Pricing Model CAPM It asserts that the expected excess return on stock i in period t should be proportional to the excess return of the market as a whole Excess returns are returns above those generated by a risk fr
4、ee asset such as t bill r it r ft i r mt r ft Note that the factor of proportionality i varies across industries possibly companies depending on how risky they are An excellent discussion of the CAPM can be found in Berndt E 1991 The Practice of Econometrics It is also discussed in your textbook on
5、page 160 Note that this relationship lends itself naturally to estimation by linear regression r it r ft i i r mt r ft i a Retrieving and Manipulating the Data 5 points The file market dta is a STATA data file containing the monthly stock market returns of a selection of companies over a 10 year per
6、iod It also contains the market return and the return on a risk free asset t bills Choose two companies from those in the data For each company create a variable for the excess returns Create a variable equal to the excess returns for the market For each of your companies plot the excess returns ove
7、r the period against the excess market return b Bivariate Regression and t tests the CAPM 15 points For each of your companies regress the excess returns over the period against the excess market return The of each company is a measure of its riskiness Which of your companies is riskier Do your resu
8、lts surprise you One of the implications of the CAPM is that the intercept i Report a t test of this hypothesis c Reverse Regression 10 points For one of your companies estimate the reverse regression of excess market returns on excess returns Does this regression estimate the same line If not why n
9、ot Compare the fit of the two regressions d Chow test for Structural Stability 15 points Another implication of the CAPM is that the Beta s should be fairly stable over time One way to test the stability of a regression parameter is a type of F test known as a Chow test Use this test to test the sta
10、bility of the for each of your companies e Multiple Regression and F tests The Asset Pricing Model 25 points A competitor to the CAPM is the Asset Pricing Model APM While the CAPM asserts that excess stock returns should only be related to the excess returns of the market the APM counters that exces
11、s stock returns may be contemporaneously related to other features of the economy Add the following data to your worksheet see the end of this pdf or use the market add xls The new variables are POIL Nominal Price of oil in FRBIND Federal Reserve Board index of industrial Production 1972 100 seasona
12、lly adjusted CPI Consumer Price Index 1967 100 Generate a variable for the real price of oil RPOIL POIL CPI Generate variables capturing the growth rate in RPOIL FRBIND and CPI note that the last will be the rate of inflation Growth rate in X X t X t 1 X t Note that the data in above starts one mont
13、h earlier than the data in market dta Generate variables capturing the surprise in the growth rates in each month by taking deviations from sample means Estimate the multiple regression of excess stock returns on excess market returns the surprise in the inflation rate rate of growth in real oil pri
14、ces and rate of growth of industrial production and Use an F test to test for the exclusion of the new variables This is a test of the CAPM against the more general APM Why Discuss your results Q2 30 points Use the data in LOANAPP DTA for this exercise The binary variable to be explained is approve
15、which is equal to one if a mortgage loan to an individual was approved The key explanatory variable is white a dummy variable equal to one if the applicant was white The other applicants in the data set are black and Hispanic To test for discrimination in the mortgage loan market a linear probabilit
16、y model can be used approve 0 1 white other factors i If there is discrimination in loan approvals against minorities and the appropriate factors have been controlled for what is the sign of 1 ii Regres approve on white and report the results in the usual form Interpret the coefficient on white Is i
17、t statistically significant at the 1 level Is it practically large Would you conclude that there is discrimination or no discrimination in the market for loans Explain iii As controls add the variables obrat loanprc unem male married dep sch and cosign Is there still evidence of discrimination again
18、st nonwhites using a 1 level of significance See below where the data is described for the definition of these variables iv Now allow the effect of race to interact with the variable measuring other obligations as a percent of income obrat Is the interaction term significant at the 1 level What do y
19、ou conclude from the hypothesis test v Re estimate the model in part iii with the inclusion of an interaction term between white and male Test the null hypothesis that the coefficient on the white male interaction term is equal to zero against a two sided alternative using a 1 significance level Are
20、 white males treated differently in the market for loans Note The LOANAPP DTA data set can be downloaded from the course Blackboard page The data set has 1000 observations and 10 columns The columns corresponds to and respectively The definition of each variable is 1 if the loan is approved 0 otherw
21、ise 1 if the applicant is white other obligations as a of income amount of loan price of the property unemployment rate in applicants industry of employment 1 if the applicant is male 1 if the applicant is married number of dependents 1 if the applicant has more than 12 years of schooling 1 if there
22、 is a cosigner Market add nth year poil frbind cpi mnth year poil frbind cpi 12 1977 7 93 126 5 166 3 01 1983 28 85 151 4 260 5 01 1978 7 90 125 9 166 7 02 1983 34 10 151 8 263 2 02 1978 7 87 127 6 167 1 03 1983 34 70 152 1 265 1 03 1978 7 79 128 3 167 5 04 1983 34 05 151 9 266 8 04 1978 7 86 128 7
23、168 2 05 1983 32 71 152 7 269 0 05 1978 7 89 129 7 169 2 06 1983 31 71 152 9 271 3 06 1978 7 99 129 8 170 1 07 1983 31 13 153 9 274 4 07 1978 8 04 130 7 171 1 08 1983 31 13 153 6 276 5 08 1978 8 03 131 3 171 9 09 1983 31 13 151 6 279 3 09 1978 8 39 130 6 172 6 10 1983 31 00 149 1 279 9 10 1978 8 46
24、130 2 173 3 11 1983 30 98 146 3 280 7 11 1978 8 62 131 5 173 8 12 1983 30 72 143 4 281 5 12 1978 8 62 133 0 174 5 01 1984 30 87 140 7 282 5 01 1979 8 50 132 3 175 3 02 1984 29 76 142 9 283 4 02 1979 8 57 133 3 177 1 03 1984 28 31 141 7 283 1 03 1979 8 45 135 3 178 2 04 1984 27 65 140 2 284 3 04 1979
25、 8 40 136 1 179 6 05 1984 27 67 139 2 287 1 05 1979 8 49 137 0 180 6 06 1984 28 11 138 7 290 6 06 1979 8 44 137 8 181 8 07 1984 28 33 138 8 292 6 07 1979 8 48 138 7 182 6 08 1984 28 18 138 4 292 8 08 1979 8 62 138 1 183 3 09 1984 27 99 137 3 293 3 09 1979 8 63 138 5 184 0 10 1984 28 74 135 8 294 1 1
26、0 1979 8 72 138 9 184 5 11 1984 28 70 134 8 293 6 11 1979 8 72 139 3 185 4 12 1984 28 12 134 7 292 4 12 1979 8 77 139 7 186 1 01 1985 27 22 137 4 293 1 01 1980 8 68 138 8 187 2 02 1985 26 41 138 1 293 2 02 1980 8 84 139 2 188 4 03 1985 26 08 140 0 293 4 03 1980 8 80 140 9 189 8 04 1985 25 85 142 6 2
27、95 5 04 1980 8 82 143 2 191 5 05 1985 26 08 144 4 297 1 05 1980 8 81 143 9 193 3 06 1985 25 98 146 4 298 1 06 1980 9 05 144 9 195 3 07 1985 25 86 149 7 299 3 07 1980 8 96 146 1 196 7 08 1985 26 03 151 8 300 3 08 1980 8 05 147 1 197 8 09 1985 26 08 153 8 301 8 09 1980 9 15 147 8 199 3 10 1985 26 04 1
28、55 0 302 6 10 1980 9 17 148 6 200 9 11 1985 26 09 155 3 303 1 11 1980 9 20 149 5 202 0 12 1985 25 88 156 2 303 5 12 1980 9 47 150 4 203 3 01 1986 25 93 158 5 305 4 01 1981 9 46 152 0 204 7 02 1986 26 06 160 0 606 6 02 1981 9 69 152 5 207 1 03 1986 26 05 160 8 607 3 03 1981 9 83 153 5 209 1 04 1986 2
29、5 93 162 1 308 8 04 1981 10 33 151 1 211 5 05 1986 26 00 162 8 309 7 05 1981 10 71 152 7 214 1 06 1986 26 09 164 4 310 7 06 1981 11 70 153 0 216 6 07 1986 26 11 165 9 311 7 07 1981 13 39 153 0 218 9 08 1986 26 02 166 0 313 0 08 1981 14 00 152 1 221 1 09 1986 25 97 165 0 314 5 09 1981 14 57 152 7 223
30、 4 10 1986 25 92 164 5 315 3 10 1981 15 11 152 7 225 4 11 1986 25 44 165 2 315 3 11 1981 15 52 152 3 227 5 12 1986 25 05 166 2 315 5 12 1981 17 03 152 5 229 9 01 1987 24 28 165 6 316 1 01 1982 17 86 152 7 233 2 02 1987 23 63 165 7 317 4 02 1982 18 81 152 6 236 4 03 1987 23 88 166 1 318 8 03 1982 19
31、34 152 1 239 8 04 1987 24 15 166 2 320 1 04 1982 20 29 148 3 242 5 05 1987 24 18 166 2 321 3 05 1982 21 01 144 0 244 9 06 1987 24 03 166 5 322 3 06 1982 21 53 141 5 247 6 07 1987 24 00 166 2 322 8 07 1982 22 26 140 4 247 8 08 1987 23 92 167 7 323 5 08 1982 22 63 141 8 249 4 09 1987 23 93 167 6 324 5
32、 09 1982 22 59 143 9 251 7 10 1987 24 06 166 6 325 5 10 1982 23 23 146 5 253 9 11 1987 24 31 167 6 326 6 11 1982 23 92 148 5 256 2 12 1987 24 53 168 8 327 4 12 1982 25 80 150 0 258 4 The new variables are POIL Nominal Price of oil in FRBIND Federal Reserve Board index of industrial Production 1972 1
33、00 seasonally adjusted CPI Consumer Price Index 1967 100 Generate a variable for the real price of oil RPOIL POIL CPI Generate variables capturing the growth rate in RPOIL FRBIND and CPI note that the last will be the rate of inflation Growth rate in X X t X t 1 X t Note that the data in above starts one month earlier than the data in market mtw
