SMI S&P500 part 2

In this second part of our ‘Single Momentum Investing’ analysis, we calculate the effect of transaction costs, the days invested and the number of transactions over time. We will also answer the question whether it is necessary to check our investments every marketday.

Transaction costs

In our first analysis of the strategy SMI S&P500 we did not take transaction costs into account. The more transactions we do, the more these transaction costs will influence the return. Transaction costs depend on the broker you are working with and the taxes in your country. In the calculation below we set the cost of each transaction at $5. The start value is raised to $5000 because we think it ’s not very wise to invest $100 if the transaction cost is $5.

With transaction costs set at $5, the CAGR of SMI S&P500 int175 over this 92-year period becomes 6,28%. It’s not onlogical that this is lower than the CAGR without transaction costs, but it remains higher than the result for the ‘buy-and-hold’ strategy, for which we have transaction costs only once.

The calculations show that in 2,24% of market days a transaction would have been done. So most of the days, we will not do any transactions; only in about 1 out of 50 market days, this strategy would make us buy or sell stock.

The days invested divided by the total number of market days is 67,34%. With this strategy in about one third of the time we would be in cash. The rest of the time we would be the proud owners of the ETF.

Controlperiods

So far we calculated the results of the Single Momentum strategy  using daily controls and, if necessary, daily selling or buying. Our readers, being active professionals, might not be able to check their investments or take action every day. Therefore, we need to evaluate whether the results are different, if we only check our investmens with regular intervals. We will call these regular intervals ‘controlperiods‘. For example, an invester who checks his investments (and buys or sells if necessary), once a week, uses a controlperiod of 5 marketdays. If you check your investments once a month (every 4 weeks), the controlperiod is 20 marketdays.

The figures below show the CAGR (%) and maximum breakdown (%) for the Single Momentum strategy applied on the S&P500 with interval 175 and for the underlying index (S&P500). As before we use data for the 92-year period from 1927 until october 2020.

CAGR% different control periods
Max Breakdown different control periods

The first thing we see when looking at the charts, is the prominent sawtooth pattern, for the CAGR as well as for the maximum breakdown. This pattern suggests there is no real optimal controlperiod, since small differences in future price evolutions can make the difference between a top and a bottom. 

We can make two valid conclusions though!

In the first place, if we look at the CAGR-chart, we see that the CAGR of the SMI-strategy remains consistently above the level of the CAGR of the ETF for controlperiods up to 11 market days. Once we use a controlperiod that’s longer than 11 marketdays, the CAGR ossilates around the reference CAGR. Even for controlperiods up to 11 marketdays, the difference in CAGR is small, but as we discussed before, over long periods of time even small differences in CAGR can make a big difference due to the compounding of returns.

Our second observation is probably more important. As we see in the second chart, the maximum breakdown for the SMI-strategy, is consistently much lower than the maximum breakdown for the index. With values of around 55% it is still very high, but much lower than the 86% for the S&P500. Needless to remember that this maximum breakdown happened during the famous bear market starting in october 1929. A limitation of the maximum breakdown is important because nobody is able to consistently predict markettops. So an investor who happened to buy at the worst moment, definetely wants to limit his losses.

Learn more in the next episode:

SMI S&P500 part 3