Since the early nineties literature regarding Momentum has significantly flourished. However, much skepticism still persists, and the value approach is still considered by many practitioners more as an alternative than the perfect complement to it. An increasing amount of academic research has presented undeniable evidence of momentum efficacy. More important, this research is backed by the founding fathers of classical finance and fundamental analysis like Fama and French, Titman and Jegadesh, and Moskowitz. A paper published in May 2014, Fact, Fiction and Momentum investing by Moskowitz, Asness and Israel, clearly outlines the results of the last 20 years of academic research on Momentum.
Trendrating is designed to complement your existing process. It can easily fit into different steps of the decision chain. It can be used to generate ideas, to validate existing positions, or to fine tune portfolio allocation. Trendrating is a set of analytics and tools which adds value to the existing decision-making process.
Analyzing momentum is extremely difficult – noise and bias are everywhere. However, momentum exists in many cases and goes unnoticed by models which only rebalance quarterly or bi-annually. Speed of analysis is crucial to promptly adjust index components. Trendrating adjusts our model components to identify price moves within 2 – 4 weeks and participates in trends for a longer duration. Our straight forward interface also enables you to create and personalize your own momentum strategies and indices without involving expensive quants or algo strategy developers.
A practical way to overlay Trendrating with your current investment strategy is to dedicate 5-10% of a portfolio to a Trendrating Momentum Strategy, using our Screener for idea generation on single instruments, countries and sectors alongside of our Index Builder to arrive at an optimal allocation within an Index or ETF. Ratings data is easily exported into spreadsheets and quantitative models for actionable workflow.
Trendrating is a service targeted to equity portfolio managers. It is designed to offer an objective metric to evaluate the quality of medium-term price trends. Trendrating is structured as a rating system that measures the true direction of the underlying trend of equities and indexes. The rating system is based on 4 grades (A= strong bull trend, B= bull trend, C= bear trend, D= strong bear trend). The rating system has been developed with the goal of capturing trends lasting from a few months to a few quarters. The rating approach provides a simple and actionable metric that generates objective trend analytics. Trendrating can easily fit into any investment decision process, bringing additional intelligence and discipline to fund management.
The Trendrating’s evaluation methods are based on genetic algorithms using multivariate data analysis on large populations of historical price patterns.
The pattern recognition is performed through numeric analysis of historical prices, volumes and volatility. Montecarlo method implements a daily Backtesting on millions of pattern series. The Backtesting evaluates the historical reliability of any single pattern and it isolates hedge cases when present. The statistical evidence of recurring pattern allows the Trendrating method to self-adjust the parameters and weightings of the 8 indicators that are at the basis of the model in reaction to behavioral changes of markets.
That numeric analysis is based on regression analysis between multi time frame performance on any security respect to of any peer in the security’s peer group.
No, the model has been extensively optimized in the past using 1,800 different configurations and we selected the most reliable and sustainable version of the model across the tested universe. Therefore the model is “written in stone” and will not be subject to further optimization. The model has a self-adaptive algorithm that works within specific boundaries of pattern recognition.
The backtesting has been performed across 20,000 time series (stocks, ETFs, indexes, currencies, commodities) for 20 years of daily data. Therefore the statistical universe for testing cover 400,000 years that offer an extremely broad variety of market cycles and individual price behaviors. The vast testing sample offer a very comprehensive universe of clean trends, choppy markets, false moves, high and low volatility, slow reversals and abrupt retracements that we believe well represent the complexity and variety of price behaviors.
The model has been developed not with the goal of maximizing results on a small set of time series, but rather to be robust across the board. By using such a large universe of historical data we avoided the risk of curve-fitting. The Trendrating model is designed to have consistency of results over time.
Trendrating has a dual purpose – bringing more discipline in the decision process and enhancing the performance. Fund managers can measure almost everything (fundamental ratios, correlation, volatility, estimates), but there is no systematic, objective, actionable way to measure price trends. Trendrating is filling this gap by adding a valuable analytical tool to complement the investment decision process. It provides a systematic assessment of price trends in a synthetic form that is easy to understand and to act upon. Even more relevant is the impact that Trendrating can have in terms of enhanced performance and reduced risks in a portfolio.
Trendrating is not a trading system and it is not a forecasting tool. It is not designed to provide entry and exit signals .It is not designed to try and forecast price moves. Trendrating measures the direction and the quality of the medium term trends. For example as long as stock has an “A” rating it is a good candidate for a buy or a hold decision, as the positive trend is intact.
Changes in the rating highlight a change in the trend. A downgrade from A to C should trigger a re-evaluation of the investment as this is often associated to a negative trend reversal.
Any professional investor in equities can benefit by using Trendrating, as long as their investment horizon is neither hours or days. Trendrating is designed to capture those price trends lasting from a few months up to a few quarters.
Trendrating covers stocks, ETFs, market indexes, sector indexes for US and international markets.
Trendrating is extremely effective at capturing the medium-to-long term trends. Trendrating statistical evidence shows that on average the rating provided has an accuracy of around 80% at identifying substantial price move greater than 5% over a prolonged period. The system has back testing functionality that allows users to test on the fly, historical results on any stock and index in the database so that anybody can grow their understanding of the model behavior and their confidence.
Prior to Trendrating, there weren’t any widely available tools to objectively measure the effects of momentum. As such, naysayers referred to it as “voodoo” and “irrational exuberance”, claiming that it followed no rhyme or reason. The recent abundance of academic research has led esteemed firms such as Blackrock, Barclays, Invesco, AQR and MSCI to launch funds and indices which have outperformed the market. We are under the strong belief that acceptance will grow as awareness continues to build.
Momentum exists in many cases and goes unnoticed by providers who only rebalance their indices quarterly or bi-annually. Speed of analysis is crucial to promptly adjust the index components. Trendrating adjusts our model components monthly so that you can identify more opportunities.
Like any strategy, momentum does not make money all the time but this goes for all factors. Momentum has a much higher historical Sharpe ratio than growth or value strategies, even when including 2009. On a risk-return basis momentum is tops. In 2009, if momentum was all you did, you had a terrible year. The same goes for 1999 for value investors. However, combining Momentum and Value effectively negates losses from sudden crashes for both practices, making their combination extremely desirable.
Frazzini, Israel, and Moskowitz (2013), scrutinized actual trades from a well-known institutional investor from 1998 to 2013. The data contained more than a trillion dollars of transactions from across 19 developed equity markets. Their conclusion was that per dollar trading costs for momentum are equally as low as other investment strategies, and despite higher turnover, momentum easily survives transactions costs.
Research from multiple well respected academic sources found that the long and short side of momentum are equally profitable using 86 years of U.S. equity data as well as 40 years of international equity data. No evidence was discovered that showed greater profits on either the short side or the long side.
Trendrating is not a market timing or trading signal system. Our algorithm is correct approximately 80% of the time at filtering out false price moves or “noise” over the course of months. Many times a stock with a high rating which has reversed direction will return to and exceed the price threshold of the original rating. Trendrating calculates the strength and magnitude of momentum for each stock it rates.
We get asked this question many times, because it would seem natural to launch a fund based on our unique and powerful model. However, we know that the fund management industry is going through a great deal of change: it’s very competitive, becoming more regulated, fees are being squeezed and costs of business are getting higher. On the other hand, Fintech is one of the fastest growing industries in the world and we believe that in 3 – 5 years Trendrating will become the undisputed world leader in momentum analytics, models, technology and services. Being selected as a strategic partner by FTSE Russell confirms our conviction. We would rather channel our efforts and resources in strategies to meet this goal.