This is what we are working on.

Existing equity indices include custodian banks, fund adminstrators and other non-asset management, general financial service companies; they are typically U.S.-centric as well. We propose an index consistening of global listed asset managers with at least 65% asset management function. Ideally such an index will allow us to observe market sentiment toward the asset management industry.

We use a combination of machine learning algorithms including recurrent neural networks together with a set of sentiment, market, and economic factors to produce forecasts up to 1 year in advance for the s&P 500. We achieve an out-of-sample r-sqaured of 0.65 or higher for some forward periods. See our white paper for more information. See the most recent forecasts here.