Chart A-5. Full-imputation index results with S5, S10, and S17 regressor sets: televisions, January 2021– June 2023
Period	S17	S10	S5
January 2021	100.000	100.000	100.000
February 2021	100.794	100.735	103.123
March 2021	101.291	101.259	107.298
April 2021	101.698	101.702	108.439
May 2021	101.620	103.082	107.607
June 2021	101.866	104.290	111.397
July 2021	107.806	111.386	120.105
August 2021	109.172	113.194	120.849
September 2021	109.143	113.597	125.873
October 2021	107.202	111.959	126.436
November 2021	101.361	105.984	120.983
December 2021	99.964	104.655	116.142
January 2022	101.244	105.808	115.727
February 2022	95.884	100.587	112.202
March 2022	94.253	99.413	110.125
April 2022	91.298	96.113	106.101
May 2022	90.666	95.778	105.651
June 2022	89.264	94.044	104.940
July 2022	90.413	95.272	106.911
August 2022	90.291	95.402	104.646
September 2022	93.131	98.624	108.810
October 2022	90.538	95.901	107.527
November 2022	83.458	88.229	99.970
December 2022	87.026	91.814	103.475
January 2023	88.230	93.064	101.982
February 2023	85.975	91.521	103.145
March 2023	86.605	91.776	107.314
April 2023	86.500	92.417	104.227
May 2023	85.191	90.420	103.919
June 2023	85.521	90.477	104.016
Note: S5 contains only binary variables for three high-value brands, screen area, and screen-area squared. S10 adds five highly predictive technological binary variables. S17 adds four less predictive technological binary variables and three more binary variables for brand. Source: U.S. Bureau of Labor Statistics.