Iscriviti alla lista d'attesa

Statistiche › Rendimenti di lungo periodo › Correlation forecasts

Correlation forecasts explain about a quarter of what later happens, and that is the best case

Correlations, 36 month horizon
0.24
Covariances, 36 month horizon
0.35
Covariances, 12 month horizon
0.18
Variances, 36 month horizon
0.52
Best variance model, 36 months
0.54

Correlation between the past sample statistic and the subsequent realised statistic, out of sample, from a study of US single stock covariances. Forecasts are formed each April from 1968 to 1997 using the prior 60 months of returns. The universe is NYSE and AMEX common stocks priced above five dollars, excluding the smallest fifth by market capitalisation.

Out of sample, predicted covariances correlate 0.16 to 0.20 with realised covariances at a twelve month horizon. Correlations are the hardest of the three moments to forecast, at about 24 per cent against 35 per cent for covariances and 52 per cent for variances.

0.16 to 0.20Correlation of predicted with realised covariances, 12 months
circa il 24%Past against future correlations, 36 months
about 52%Past against future variances, 36 months
0.36 to 0.54Forecast regression slope, 12 months

Solo per membri

Il resto di questa pagina è per i membri

Below this point there are 5 sections, 1 table and 4 named sources, roughly 600 words of it. Every figure carries the source it came from and the date the data is from.

Puoi continuare a consultare gratuitamente ogni statistica della libreria. L'introduzione e il riepilogo sono sempre accessibili.