Tuesday, February 07, 2012

A Black Box Approach to Portfolio Performance Measurement

Dear readers,

I have been asked questions like - What is your portfolio's CAGR? What is the cost price of the stock that you have purchased? When did you buy the stock? etc.

Frankly speaking, I had not been able to answer most of these questions. In fact, I don't really compute CAGR or look at the cost price of each stock in my portfolio. I don't even remember when did I bought a stock, especially for those that had been purchased years ago. Then how do I measure the performance of my portfolio? This post attempts to answer this question.

I like to use Engineering analogy when managing my investment portfolio. Because I am an Engineer by training, sometimes I don't really like to use financial terms to analyze or describe stuff. Therefore, you don't see spreadsheets, excel tables, detailed analysis on companies etc on my blog here. I am not an analyst and I will not attempt to be one here. I am also not an accountant and therefore I do not claim to be an expert in numbers.

What I can do though is to try to put everything together and make it work. An Engineer will not claim that he understand every component in a system and how they behave, but at least he try to understand how the system behave in normal/adverse condition and make small changes based on observations. But there will not be a drastic re-design to a system that is working most of the time.

A black box approach to portfolio performance measurement is what I had been doing most of the time. What it means is that without looking at what is the content of the portfolio, I just get the closing market value of my portfolio every day to have a feel of how my portfolio as a whole is behaving. From there, I can roughly know how my portfolio react to market movements every day. Is it volatile? Has it underperformed? etc. If my portfolio is behaving normally most of the time based on my observation, I will not go into individual stock at all. Which means, I will not bother looking at the closing market price of each stock in my portfolio, let alone decide which stock to sell. A black box approach is a simple yet effective method because once you select good stocks to invest, your portfolio should be stable enough to withstand market downturn. Granted, there are some stocks which might not perform to your expectation after you introduced them but as long as your portfolio on a whole is well-diversified and you keep on over-weighting the better ones, it should be good enough. You should not attempt to make changes on a portfolio that is working most of the time, just like you don't do a re-design of a system that is working and already been deployed to be used by customers. Yes, the system is not perfect and there is some bugs but as long as it works most of the time, it is good enough. Learn to live with some junk stocks in your portfolio.

Another simple method I did is to compute the dividend yield of my portfolio. This is obtained by using the year-end CDP statement which detailed all the dividend that had been received during the year. Again, it is just simply adding together all the dividends and derive the yield based on my portfolio size at the end of the year.

Time should be spent on looking at good companies to invest at the right price rather than monitoring the share price of each stock in your portfolio or computing returns, which is really the result rather than the process. Focus on the process and the result will come automatically.

Labels: ,


Blogger Cory said...

Hi GH Chua, what do you base on to determine your process of selection is working.


7:10 AM  
Blogger ghchua said...

Hi Cory,

Getting data points using the black box approach is one way to determine whether the system (or in this case my portfolio) is working. If the system is working, then obviously the process of selection is working as well. If your selection process is not working, the stocks in your portfolio will cause the data points that you have collected to signal that it is unstable.

Of course, an unstable output can be due to several factors, and sometimes it is just market condition. But if the market is doing ok overall and your portfolio fluctuates like mad every day, then something is wrong and you can start to investigate the problem.

But what I want to say in this post is really don't overreact and start to switch stocks in your portfolio every single moment. Sometimes, it is better to just leave it if there is no major problem.

7:56 PM  
Blogger cif5000 said...

Have you ever wondered that your black box is working well because it is placed in a right environment? Throw it into another market, the output will be vastly different even if you apply the same approach.

See you soon in the coming AGMs...

10:35 PM  
Blogger ghchua said...

Hi cif5000,

Well, what you said can be true. But we will never know unless we put the portfolio under the respective markets and collect data points based on it.

But we must be careful not to do things differently. Because my portfolio consists of SGX listed companies, I will not want to benchmark against other markets around the world. It is similar to what fund managers are doing. A fund that is invested in SGX listed companies would not want to benchmark against, say NASDAQ.

See ya soon in the next AGM! :)

1:15 AM  

Post a Comment

<< Home


A self-directed investor, looking to invest for retirement needs and bypass all those expensive financial planners/insurance agents. Investing is fun, profitable or most important of all, knowledge gained is useful for the rest of your life!

Powered by Blogger