RELATION TO MONEY AND PRICES

comparing these series over the period 1915-1927 arehigh; without lagging r = + 0.709; lagging one quar-ter, r = -f 0.829; two quarters, r = -f- 0.891; four quar-ters, r = -f 0.838. In both periods the coefficients ofcorrelation grow smaller as the P’s are lagged behind i,while they grow larger when i is lagged behind P. Theresults from the analysis of the short term data, whilediffering in some respects, may be said to confirm the re-sults obtained from comparing the long term data.

§9. Elimination of Trends

These high correlations do not necessarily mean thatthe interest rate will always be high when prices arehigh and low when prices are low, but the tendencytoward this is definitely established.

The correlations obtained for all periods and sub-periods considered are unusually high. It is necessary toguard against the possibility that these coefficients areof the familiar nonsense type, and are spuriously highbecause of the presence of secular trend forces that affectboth P and i. Due consideration was given to the controldevices that have wide acceptance in the literature ofstatistics, such as “elimination of trend” and “seasonal”fluctuations. The general methodology of analysis of timeseries is still in the process of formulation. The specificproblem of trend analysis is still largely unsolved. 15 Inthe present case, it is rather doubtful that trend forces areinvolved which should be eliminated. What is desired in

“See Pearson and Elderton, The Variate Difference Method, Bio-metrica, Vol. XIV, 1923, pp. 281-309; W. M. Persons, Statistics andEconomic Theory , Review of Economic Statistics, Vol. VII, No. 13,July, 1925, pp. 179-197; Anderson, The Decomposition of StatisticalSeries into Components, Journal of the Royal Statistical Society,Vol. XC, pt. 3, 1927, pp. 548-570.

[431]