ETF Trading – When to Trade Our Signals

Written by on

As we discussed in a previous post (and the FAQ), our systems show results for trading at the Monday (or first trading day of the week close). We’ve been asked a few times about whether it is strictly necessary to trade at or near the close on Monday. Over the years we have done a lot of testing of various parameters for market timing, ETF trading and stock rotation strategies.

One of the things we have studied extensively is what time at which it is best to trade our systems. The systems are designed to be traded at the close on the day after a signal, but does it make a difference if we trade at, say, the opening of the next day after that? Our backtesting has indicated that for the systems there is not strong advantage to trading exactly at Monday’s close.

Going back, using a few years of data, the differences between trading at Monday’s close or Tuesday’s open are not significant – although, obviously, week-to-week, there are variations, the differences seem to even out over time if trading is done consistently at Tuesday’s open. In fact, we are so unconcerned with the exact time of trading, that we often trade using the morning window trade at FolioFn (around 11 AM) – and we use the Monday morning window.

TSP Timing – Why Only the S, I, and F Funds?

Written by on

Sometimes we are asked why our TSP timing systems only trade between the S (US small cap), I (international stocks) and F (bond) funds. When we were developing our first TSP timing signal, it didn’t take long to come to the conclusion that these were the only funds worth concentrating a TSP trading strategy on. This was because when developing the trading system, a close look at the historic record for these funds and their stock fund analogs showed that when it was time to be aggressively “in” the stock market, it was far better to be invested in small caps or international stocks than to be in the large cap US segment (the “C” fund).

Switching between small caps and international stocks provided a much greater return than staying in large caps. So, if we are ready to be aggressive in our approach, it is better to move into the funds that will reward our risk appropriately. Similarly, although the G fund (a money market equivalent) seems on the surface to be the best place to be when taking a conservative “out” of the market position, an examination of the historic returns showed that the F fund provides better returns when out of stocks with only a small amount of additional risk. When we developed the TSP systems, our early efforts attempted to bring all of these funds into play.

However, placing the C and G funds into the mix did very little to reduce risk, but definetely reduced the overall returns – and did so dramatically. This is why the systems trade only the S, I and F funds – the risk-reward ratio for the systems did not improve enough when adding the C and G funds to the trading mix.

ETF Trading System Results (Applies to the TSP Systems as Well)

Written by on

We just posted a new FAQ (the FAQ link is also available at the bottom of each page of the site).  We are trying to answer a few common questions:

When do the ETF and TSP systems trade? How do you calculate the week to week results? Why does the system report a gain for the week when the new trade is down?

The systems trade at the close of the first trading day of the week (usually Monday). This means that when a system begins a new trade, the old trade is closed out at the Monday close and the new one is entered at the Monday close. Since we calculate the returns and post the signals on the weekend, the system returns are calculated on a week to week basis – Friday close to Friday close.

Thus, on a week where there has been a new trade, the weekend update takes into account what the old trade did on the Monday, and what the new trade did on Monday close through the Friday close. This explains why it is possible that the system will show a gain for the week even though the new trade, as of the Friday close, is down. Of course, the converse is sometimes true – the system will show a loss for the week even though the new trade is up for the week.

To summarize – the system results we post take into account the performance of the old trade on the Monday when a new trade is entered. The system results show week-to-week results calculated Friday-to-Friday even though the trades are closed and entered on Monday at the close.