A Winton research project with UC Berkeley used a technique for adding noise to datasets that makes it impossible to identify individuals, while preserving broader statistical patterns.
Our research explores alternative approaches for assigning sectors to the S&P 500 using stock price covariance and natural language processing.
Equities and bonds are often assumed to be negatively correlated. This hasn’t always been true.
Market shocks may feel like they are becoming more common, but the data suggests otherwise.
Does globalisation mean that stock returns are related more to sector than country performance?
October has been the most volatile month for stocks on average over the past 87 years. Is this due to chance?
We find hidden costs of about 10 basis points per year for the naïve global index tracker.
When investing in multiple hedge funds, investors eventually face the challenge of finding an effective rebalancing methodology.
A unique experiment allowed us to demonstrate the effectiveness of our research process in a relatively short time.
Publication bias could have a part to play in the disappointing performance of popular equity indicators.
We put the recent fall in US equity volatility into context.
Adding a new, uncorrelated strategy to a portfolio can improve the risk-adjusted performance of the original portfolio.
We caution against drawing conclusions from short time series and emphasise the importance of acknowledging the uncertainty on performance estimates.
Without an appropriate understanding of data, it can be too easy to discover fool’s gold.
Often tracking error is incorrectly defined ‒ with important consequences.
Randomly selected global equity portfolios have outperformed market-capitalisation-weighted portfolios over the past 15 years.
Roll yield, a source of profits for trend followers, is the return captured when a futures contract converges to the spot price.
The long-term correlation of trend following with equities is effectively zero. But there are many misconceptions regarding what this means.
Ignoring uncertainty in portfolio construction can lead to a large gulf between the realised and expected performance of a strategy.
Analysis of the correlation structure between futures markets, S&P 500 stocks and trend-following strategies over the past 50 years.
We estimate a hidden turnover cost of about 20 basis points per year for investors tracking the S&P 500.
We find evidence that hypothetical performance data can be significantly over-optimistic compared to subsequent realised performance.
"Meta-overfitting" may be endemic in finance as well as other fields of research.
We consider if trend-following performance is dependent on trading speed.
The determination of good from bad investments using quantitative measures.