APAC Insider Transactions Strategies Overlaid with Sentiment

RavenPack | July 07, 2020

How can investors use insider transactions overlaid with sentiment data to drive alpha and what are the results if applied to portfolios from the Asia Pacific region? RavenPack’s latest white paper investigates.

In an earlier white paper, RavenPack’s data science team developed a strategy using Insider Transactions and news sentiment and found it to provided excess market returns of 8.6% and 13.9% for global Mid/Large-Cap and Small-Cap stocks, respectively.

In their latest white paper, they apply a similar methodology to stocks in the Asia Pacific (APAC) region.

Asia Pacific Insider Transactions Data

Whilst preceding studies proved incorporating Insider data could be profitable on a global scale, increased interest from clients in the APAC region led data scientists at RavenPack to ask whether the same would be said of strategies using only portfolios of Asia Pacific stocks.

Could trading a portfolio of APAC stocks solely using insider buy and sell signals generate excess returns, and could those returns be enhanced by applying a news sentiment overlay?

The first step was to use the insider data to develop an indicator that could provide the basis for a trading strategy.

RavenPack Insider Transactions data includes every type of insider trade made by every type of insider, yet to ensure good signal strength, it was decided that only those trades executed by mid-to-higher level executives with a significance rating of at least ‘3’ should be counted.

In effect, this meant only including regular buy and sell trades excluding “non-intentional and mechanical transactions, such as awards of shares, exercises of options, tax-related transactions, remuneration, and share plan purchases.”

The resulting strategy went long or short stocks depending on the relevant insider transactions’ signal.

Backtesting was run for signals aggregated over different periods, including 1,5,10, and 21 days. Trades were closed following the generation of an opposite signal.

The longer signal aggregation periods tended to generate larger portfolios because of the scarcity of signals. This explains why the 21-day aggregation window in the table below includes the largest portfolio of stocks - there is more chance an insider transaction providing a signal occurred within a 21-day window than over a 5-day period.

Asia-Pacific portfolio strategy stats

As can be seen from the stats above the most profitable strategy was to use an aggregation window of 1-day. This yielded an annual excess return of 7.9% for mid-large cap stocks and 11.8% for small-cap stocks, and an information ratio of 1.65 and 1.76 respectively. At the same time, a longer aggregation of 21 days still produced a positive 2.5% and 6.9% respective return.

An analysis of short and long trades found the former to consistently outperform the later, suggesting one optimization strategy might be to weight the strategy in favor of short trades.

Performance decay for long and short legs of the Asia-Pacific Universe portfolios

To ensure the strategy’s returns were not attributable to other factors, its performance was tested using the MSCI Barra GEM3 model to decompose insider signals from other market factors. It turned out the strategy’s performance was found to be very similar across all aggregation periods and market capitalizations, demonstrating persistent alpha generation.

Sentiment Overlay

A news sentiment overlay helped improve the strategy performance, especially on the short side, with insider sell signals following a 3-month period of negative news sentiment tending to improve the quality of the subsequent short trade.

Price reaction to insider trading under the influence of news sentiment



“We can generalize that companies faced with negative media attention before significant insider net sells are likely to witness greater underperformance than in the absence of this negative trend, which possibly indicates a confirmation to the broader market that the insider may not be expecting any changes in the company’s performance,” says Peter Hafez, Chief Data Scientist at RavenPack.

Easily implement the results of this White Paper using the RavenPack Analytics Platform, which includes a comprehensive and high-quality database of global insider transactions. Request a trial here.

Request White Paper

Request a Trial

Fill out the form below and see RavenPack in action.