Other ETF Options

In my previous posts we have talked a lot about the ETF SPY. Today I am going to be discussing to new ETF's, GLD and QQQ. GLD is a relatively cost efficient and secure way to access the gold market. It is the largest physically backed gold ETF on the market. GLD shares may be purchased from the trust only in one or more blocks of 100,000 shares. The shares are sold at a price that is determined by the price of gold at the time and the trading price of the shares. Meanwhile, QQQ is an ETF that includes 100 of the largest international and domestic companies listed on the Nasdaq stock exchange. QQQ's stock holdings are dominated by the big-tech companies such as Apple and Google. Shares of QQQ can be purchased on the New York Stock Exchange(NYSE).

In this discussion, I am going to compare SPY, QQQ, and GLD. To do this I will look at the average monthly returns, standard deviation, and Sharpe ratios. For SPY these are 1.20%, 4.44%, and 24.89% respectively. The numbers for GLD are 0.o6%, 4.50%, and -0.86% respectively. Lastly, for QQQ the data is 1.69%, 3.61%, and 43.97% respectively. When looking at these numbers QQQ is the preferred ETF. QQQ has the highest average monthly return with the smallest standard deviation as well as almost double the Sharpe that SPY does! This tells me that QQQ would be a very meaningful investment.

Now, I will go on to estimate the beta for the two new ETF's and calculate Treynor's measure as well. To do this I will use the "linest" function in Google Sheets. So, while using SPY as the market portfolio the estimated beta for QQQ is 0.247 and for GLD it is 0.064. With these beta values it is assumed that both stocks are not very risky, but that QQQ is riskier than GLD. However, you can also expect to get a higher return from QQQ as we saw previously when we calculated the means for each ETF.

The Treynor's measure takes into account annual return over beta. This measure for SPY, QQQ, and GLD are 0.5848, 0.8195, and 0.1150 respectively. QQQ again looks to be the better ETF as it has the highest Treynor's measure.

I will now create a couple of portfolios with different scenarios. In the first portfolio, Portfolio X, I will mix GLD and QQQ into a portfolio with a desired beta of 1.5. The expected return of this portfolio came out to be -1.02% with 166% invested into GLD and -66% invested in QQQ. The other scenario, Portfolio Y, still using GLD and QQQ, but this time I want a monthly expected return of 1%. The beta of this portfolio is 0.569 with 43% invested into GLD and 57% invested in QQQ. The Treynor's measure for Portfolio X is -0.08 and for Portfolio Y it is 0.208. Therefore, based on the Treynor's measures alone of SPY, GLD, QQQ, Portfolio X, or Portfolio Y, I would recommend to my clients to invest in QQQ.

Finally, I will look at whether the Oracle of Omaha has earned abnormal returns in running Berkshire Hathaway (BRK). The alpha for BRK is 0.0088 and the beta is 0.2222. To determine whether Buffett's reputation is well justified I will use the information ratio. To calculate the information ratio I must first create a tracking error series. The average and standard deviation of the tracking error series is 0.0095 and 0.0337 respectively. Therefore, the information ratio is 0.2822. The range that is considered good for the information ratio is 0.4-0.6, so in this case BRK is not considered to be superior to SPY and QQQ. So, Warren Buffett's reputation when using these numbers is not well justified. ETF's such as SPY and QQQ have been doing so well that BRK's returns are not abnormally larger than SPY or QQQ. The alpha is also low so that means the probability of the results being incorrect is not very likely.