Health	Social	Income	College
52	58	80	0
55	68	43	0
80	66	284	0
93	67	159	1
55	94	70	0
92	65	270	1
94	59	288	1
81	96	57	1
60	96	124	0
97	85	219	1
52	84	135	0
71	93	198	0
89	82	307	1
65	95	173	0
61	89	262	0
65	59	219	0
74	61	212	0
98	72	287	1
89	95	273	1
63	71	189	0
93	72	195	1
90	79	257	0
53	84	21	0
94	75	137	1
88	64	220	1
94	73	161	1
73	76	162	0
52	87	71	0
53	71	213	0
94	97	314	1
91	71	305	1
92	75	187	1
86	58	274	0
55	79	177	0
95	92	185	1
88	83	123	1
90	70	320	1
91	81	291	1
98	60	305	1
66	71	89	1
59	58	36	0
96	63	241	1
51	80	131	0
77	86	193	0
57	63	74	1
55	69	77	0
58	61	59	0
52	81	49	0
81	93	113	1
52	96	117	0
72	88	248	0
93	88	213	1
94	74	278	1
76	75	143	1
54	97	22	0
91	61	205	1
97	82	183	1
81	89	255	0
65	64	69	1
56	66	211	0
81	92	137	1
55	82	21	0
64	64	194	0
88	63	157	1
65	63	138	0
91	85	175	1
63	65	103	1
75	80	119	1
55	64	136	0
57	63	25	1
53	62	24	0
53	74	96	0
58	77	47	0
53	74	82	0
55	84	97	0
91	80	271	1
96	82	140	1
70	76	220	0
95	70	282	1
92	71	227	1
53	84	138	0
53	69	31	0
52	64	169	0
62	71	198	0
92	94	207	1
92	72	232	1
91	75	297	1
92	89	281	0
95	86	202	1
55	63	235	0
92	97	303	1
81	64	216	1
91	98	226	1
89	94	202	1
93	76	166	1
50	64	31	1
96	94	205	1
74	90	118	1
98	79	318	1
51	66	105	0
94	96	171	1
89	69	185	1
59	97	131	0
63	84	187	0
59	92	34	0
93	91	305	1
92	73	307	1
88	94	283	1
58	58	69	1
59	76	93	0
90	71	126	1
86	93	96	1
94	94	148	1
53	78	66	1
73	89	38	1
60	74	78	0
96	79	296	1
96	77	178	1
66	67	90	1
58	90	35	0

The accompanying data file contains survey Information on 120 American adults who rated their health (Health) and social connections (Social) on a scale of 1 to 100 . The file also contains information on their household income (Income, in $1,000 s) and college education (College equals 1 If they have completed a bachelor's degree, 0 otherwise). Consider two linear regression models for Health. Model 1 uses Soclal, Income, and College as the predictor varlables, whereas Model 2 also Includes the interactions of Soclal with Income and Soclal with College. Click here for the Excel Data Flle a-1. Use the holdout method to compare the predictability of the models using the first 90 observations for training and the remaining 30 observations for validation. Report the estimates of Models 1 and 2 derlved from the training set. Note: Negatlve values should be Indlcated by a minus sign. Round your answers to 2 decimal places. a-2. Calculate the RMSE of the two models in the validation set. Note: Do not round Intermedlate calculatlons and round your final answers to 2 decimal places. a-3. Which model is better for making predictions? b-1. Use the k-fold method, with k=4, to calculate the average RMSE of the two models. Note: Do not round Intermedlate calculatlons and round your final answers to 2 decimal places. b-2. Is the preferred model the same as found by the holdout method? Yes, because the average RMSE equals the RMSE with the holdout method. No, because the average RMSE does not equal the RMSE with the holdout method. Yes, because the earlier preferred model also has the lowest average RMSE. No, because the earlier preferred model does not have the lowest average RMSE