There are several conditions for applying equal weights as an alternative to least squares weights. Psychometric parallelism, one of the conditions, has been suggested as a necessary and sufficient condition for equal-weights aggregation. The purpose of this study is to investigate the effect of psychometric parallelism among predictors on the efficiency of equal weights and least squares weights. Target correlation matrices with 10,000 cases were simulated so that the matrices had varying degrees of psychometric parallelism. Five hundred samples with six ratios of observation to predictor = 5/1, 10/1, 20/1, 30/1, 40/1, and 50/1 were drawn from each population. The efficiency is interpreted as the accuracy and the predictive power estimated by the weighting methods. The accuracy is defined by the deviation between the population R² and the sample R² . The predictive power is referred to as the population cross-validated R² and the population mean square error of prediction. The findings indicate there is no statistically significant relationship between the level of psychometric parallelism and the accuracy of least squares weights. In contrast, the correlation between the level of psychometric parallelism and the accuracy of equal weights is significantly negative. Under different conditions, the minimum p value of χ² for testing psychometric parallelism among predictors is also different in order to prove equal weights more powerful than least squares weights. The higher the number of predictors is, the higher the minimum p value. The higher the ratio of observation to predictor is, the higher the minimum p value. The higher the magnitude of intercorrelations among predictors is, the lower the minimum p value. This study demonstrates that the most frequently used levels of significance, 0.05 and 0.01, are no longer the only p values for testing the null hypotheses of psychometric parallelism among predictors when replacing least squares weights with equal weights.

This study is a thesis ending a 120 credit masters program in Mathematics with specialization Financial Mathematics and Mathematical Statistics at the Royal Institute of Technology (KTH). The subject of Smart beta is defined and studied in an index fund context. The portfolio weighting schemes tested are: equally weighting, maximum Sharpe ratio, maximum diversification, and fundamental weighting using P/E-ratios. The outcome of the strategies is measured in performance (accumulated return), risk, and cost of trading, along with measures of the proportions of different assets in the portfolio. The thesis goes through the steps of collecting, ordering, and ”cleaning” the data used in the process. A brief explanation of historical simulation used in estimation of stochastic variables such as expected return and covariance matrices is included, as well as analysis on the data’s distribution. The process of optimization and how rules for being UCITS compliant forms optimization programs with constraints is described. The results indicate that all, but the most diversified, portfolios tested outperform the market cap weighted portfolio. In all cases, the trading volumes and the market impact is increased, in comparison with the cap weighted portfolio. The Sharpe ratio maximizer yields a high level of return, while keeping the risk low. The fundamentally weighted portfolio performs best, but with higher risk. A combination of the two finds the portfolio with highest return and lowest risk. / Denna studie är ett examensarbete som avslutar ett 120 poängs mastersprogram i Matematik med inriktning mot Finansiell Matematik och Matematisk Statistik på Kungliga Tekniska Högskolan (KTH). Ämnet Smart beta studeras i kontexten av en indexfond, där de olika testade principerna för viktning i portföljerna är: likaviktad, maximerad Sharpe-kvot, maximerad diversifiering, och fundamental viktning användandes av P/E-tal. Utfallet i testerna utvärderas i ackumulerad avkastning, portföljrisk, kostnad att handla i portföljen, och ett antal mått på fördelningen av tillgångarna. Studien går stegvis igenom processen för att samla in, ordna, och ”tvätta” data. En kort förklaring av historisk simulering, metoden för att estimera stokastiska variabler såsom kovariansmatriser, är inkluderad, såväl som en analys av distributionen av data. Processen för att optimera portföljerna och hur regler för att vara en UCITS-fond kan omformas till optimeringsvillkor beskrivs. Resultaten indikerar att alla utom den mest diversifierade portföljen har högre ackumulerad avkastning än den marknadsviktade portföljen under testperioden. I alla testade fall ökar handelsvolymen liksom marknadspåverkan när en annan strategi än marknadsviktad används. Portföljen med maximerad Sharpe-kvot ger en hög avkastning med bibehållen låg risk. Den fundamentalt viktade portföljen ger bäst avkastning, men med en litet förhöjd risk. Kombinationen av de båda metoderna ger den portföljen med högst ackumulerad avkastning och samtidigt lägst risk under testperioden.