In the Survey of Current Business, the U.S. Department of Commerce publishes data on farm commodity prices. Given are the cotton prices from November of year 1 through February of year 4.The prices are indexes with a base of 100 from the period of 1910 through 1914.Use these data to develop auto-regression models for a 1-month lag and a 4-month lag. Compare the results of these two models. Which model seems to yield better predictions? Why?

Time Cotton
Period ......... Prices
November (year 1) .... 552
December ....... 519
January (year 2) ...... 505
February ........ 512
March ......... 541
April ......... 549
May .......... 552
June ......... 526
July .......... 531
August ......... 545
September ........ 549
October ........ 570
November ........ 576
December ........ 568
January (year 3) .... 571
February ....... 573
March ....... 582
April ....... 587
May ........ 592
June ........ 570
July ........ 560
August ....... 565
September ...... 547
October ....... 529
November ....... 514
December ....... 469
January (year 4) .... 436
February ....... 419