Question

PE	BETA	EARN	DIV
12.64	0.466	0.169	-0.013
10.8	0.545	0.016	0.014
19.9	0.597	0.066	0.05
11.3	0.651	0.021	0.014
13.27	0.678	0.071	0.006
13.71	0.697	-0.004	-0.008
17.61	0.781	0.054	0.051
15.97	0.795	0.12	-0.015
10.68	0.836	-0.001	-0.021
9.63	0.845	-0.018	-0.011
16.07	0.851	0.077	-0.008
16.2	0.865	0.051	0.074
17.01	0.866	0.068	0.11
13.42	0.869	0.063	0.018
8.98	0.894	-0.137	0.007
14.49	0.898	0.062	0.044
17.84	0.906	0.12	-0.011
12.15	0.916	-0.01	0.004
11.82	0.931	0.01	0
16.08	0.944	0.081	0.045
16.95	0.952	-0.043	-0.005
12.06	0.958	0.028	-0.015
17.16	0.965	0.06	0.048
16.1	0.973	0.045	0.047
13.49	0.979	0.079	0.079
28.28	0.983	0.023	0.009
16.33	0.986	0.028	0.025
12.25	0.993	0.129	0.13
11.26	0.995	0.034	0
14.16	0.996	0.099	0.001
11.47	1.008	0.186	-0.011
17.63	1.019	0.052	0.062
14.66	1.022	0.025	0.058
14.24	1.024	0.045	-0.018
11.86	1.048	0.038	0.019
16.11	1.064	0.049	0.086
14.43	1.073	0.073	-0.011
9.42	1.075	0.062	0.02
15.25	1.081	0.042	0.026
15.16	1.091	0.051	0.015
17.58	1.097	0.114	0.098
23.29	1.122	0.066	0.072
13.14	1.133	0.094	-0.003
12.99	1.136	0.01	0.021
18.18	1.137	0.112	0.06
18.94	1.147	0.082	0.046
16.93	1.15	0.051	0.052
8.36	1.184	0.019	0.008
17.52	1.196	0.047	0.044
11.21	1.206	0.052	-0.023
16.6	1.213	0.067	0.025
10.64	1.216	0.137	0.001
10.26	1.222	0.069	-0.011
12.41	1.256	0.03	-0.014
19.33	1.258	0.052	-0.013
9.06	1.285	0.056	-0.023
21.68	1.29	0.085	0.055
11.93	1.306	0.169	0.017
8.62	1.308	0.016	0.026
9.58	1.454	0	0.005
12.02	1.464	0.022	0.012
11.32	1.525	0.05	0.006
9.92	1.63	0.038	0.018
11.75	1.821	0.112	0.05
12.41	1.857	0.108	0.037

The data is a cross-section of 65 companies averages in the S&P 500 over 1996-2000, consisting of price/earning ratio (PE), dividends paid (VID), earning growth (EARN), and a measure of company riskiness (BETA).

1. Run a regression of log of PE on EARN DIV and BETA employing a semilog(in Y) functional form. What is the interpretation of the coefficient estimates? Do you think this functional form is appropriate for the relationship of PE to EARN, DIV and BETA?
2. Try a double-log as an alternative model. Comment on the results. Which model seems more appropriate?
3. Is there any evidence that EARN may be an irrelevant variable?