English
Sett deg på ventelisten
Bøkene... men ikke gratis
10 Strategier Som Slår BörsenSvenska10 Strategier Som Slår Börsen10 Strategies That Beat the MarketEnglish10 Strategies That Beat the MarketProRealTime Coding Made SimpleEnglishProRealTime Coding Made Simple
600+StatistikkMarkedsstatistikk og edger1,327StrategierBibliotek med testede strategier18ForskningsrapporterStudien bak algoene301DokumentasjonProBuilder-referanse27AlgoutviklingsprogramKurs og leksjoner24Alt dette mater de 24 algoeneSe algoene

Statistikk › Backtesting og evidens › Liquidity anomalies

Ninety three per cent of published liquidity anomalies fail replication

Illiquidity, one month hold
1.31
Illiquidity, six month hold
1.73
Illiquidity, twelve month hold
1.99
Share turnover, all horizons
0.60

t statistics on the high minus low decile return for liquidity based sorts, January 1967 to December 2014, monthly deciles formed on exchange based breakpoints with value weighted returns. The twelve month figure of 1.99 is described in the paper as marginally significant. The turnover row is an upper bound: the paper reports that its returns of minus 0.1 to minus 0.15 per cent a month are all within 0.6 standard errors of zero.

Of 102 trading friction variables re run with value weighted returns and exchange based breakpoints, 95 are insignificant at the five per cent level. The illiquidity measure itself returns 0.28 per cent a month with a t statistic of 1.31 at a one month holding period.

95 of 102, or 93%Trading friction variables that fail at 5 per cent
0.28% a month, t = 1.31Illiquidity measure, one month holding
0.42% a month, t = 1.99The same, twelve month holding
within 0.6 standard errors of zeroShare turnover, all horizons

Kun for medlemmer

Resten av denne siden er for medlemmer

Below this point there are 4 sections and 3 named sources, roughly 600 words of it. Every figure carries the source it came from and the date the data is from.

Du kan fortsette å se på all statistikk i biblioteket gratis. Innledningen og sammendraget er alltid åpne.