The corpus obtained when buying “low” or when the Nifty is below its 200 day moving average (dma) is compared with the corpus from buying “high” (nifty above 200 dma). Apologies for yet another post on this subject, but I had to satisfy my curiosity. Kindly bear with me.
Before we begin a clarification on nomenclature. As we all know, SIP is systematic investment plan. All thismeans, is investing with a system in place. Buying each month is just one of those systems. In the previous posts, I had used terms like low-SIP and high-SIP. This simply means, buying when the Nifty is below or above a certain average (10 month was used). The observation is made once a month and investing is done only the condition is satisfied.
In this post, I shall replace the 10-month moving average with the 200 day daily moving average. Observations are done daily. And as far as this post is concerned, there are three types:
Buying low or < 200 dma: Once Nifty crosses below (<) the 200 dma value, investing is done. And it is done on all days the Nifty is below the 200 dma. In an analysis like this I cannot pick and choose which “low” day to invest in. So it is better the risk is spread by investing on all low days. Buying high or > 200 dma: Once Nifty crosses above (>) the 200 dma value, investing is done. And it is done on all days the Nifty is above the 200 dma.
Normal daily SIP: Investing is done daily. Not very practical, but I did it just for kicks. Not much will be made out of this. We will focus on the above two.
So I use the Nifty price index from Jan 1999 to Nov 2016. Those who understand what statistically significant means and value it, will agree that this is a short time period, especially for longer durations like 10Y and 15Y.
There are 3062 5Y periods between those two dates; 1819 10Y periods and 575 15Y periods. This is not a lot, but for it is worth, here it goes. If anyone is interested, I will be happy to repeat this study with the S&P500 index where daily data is available from about 1967 or so if memory serves me right.
As in the previous post, I shall assume that cash kept away for investing will earn 8% a year or about 0.021% a day. I assume that is earned only on business days. So when the time is ‘right’ or when the nifty is above or below the 200 DMA, the money put away + interest earned will be invested.
The total investment made in both buying high and buying low is either above (due to interest) or equal to a daily SIP.
FIVE YEARS
Buying “low” is compared with buying “high”. The corpus obtained over 5Y of buying low (< 200dma) and buying high (> 200 dma) is shown below.
Notice that the buying high has generated a higher corpus for many periods. There are 3062 data points in each of the above curves and they represent a 5Y period.
The excess corpus from buying “low” and excess investment from buying “low” is plotted below.
Notice that the excess corpus (blue) is negative. This means that there is no excess!! Buying low has resulted in a lower corpus for many periods except during the 2008 crash. Of course the difference is small. The point however is, buying low is not superior and buying high is not inferior! And by extension buying daily is okay too. And as established before there is not much difference between buying daily and buying month without reference to index value.
Sometimes buying low has a lower total investment than buying low and sometimes higher. Not by much though. Sometimes buying low results in a lower corpus for a higher investment!
I hope the manner in which I have plotted the comparison is clear. Apologies if it is not.
In a previous study on “does market timing work”, the key lesson was: Sometimes it does and sometimes it does not. There is no way of knowing this in real-time. Hindsight like this study is a different matter of course.
One can argue that buying low will win in the future. I have no issues with that. My goal is to convince myself (not you) that buying high is okay too and so is buying ‘whenever’. And I am convinced.
One can also argue that an investor can actively keep track of trends avoids losses and make only gains. I am not intelligent enough to respond to that. All I know is, if I am going to walk in the rain, I cannot hope to remain dry by walking between the raindrops. And I speak only for myself.
For the record, the daily SIP corpus is plotted with the other two below.
The pattern seen for 5 years is going to repeat over 10Y and 15Y, but with much lower data points as mentioned before. So I am going to present the graphs with no further commentary.
TEN YEARS
FIFTEEN YEARS
MDBSC*
My Dull and Boring SIPs Continue. The * stands for risk management with the intended financial goal in mind.
YOU can be Rich too with goal based investing
Our new book helps you answer the right questions on money management and invest with a clear purpose! Now available as hardcover and on Kindle.
Including 9 online calculator modules to create your own financial plan.
By PV Subramanyam and M. Pattabiraman.
Hardcover (₹ 375)
Kindle (244.30)
Pattu Sir,
Just a few thoughts.
1. If we are following the 200 DMA based approach as against SIP, then we would invest cash accumulated till then as soon as Nifty goes below 200 DMA and not every day that Nifty is below 200 DMA right? In this post what is the assumption made?
2. Also, if market timing is the objective, then every time Nifty goes above 200 DMA, then you would also redeem and invest back only when it goes above 200 DMA and your exit load period is over. While in practise this may be cumbersome, (not to mention transaction costs) mathematically, how does that pan out?
3. Is there a base excel sheet (Like your other calculators) from which these graphs are derived?
On a different note, much as I try I am unable to use the direct equity calculator that you had posted some time back. I did log in into value research and then try the sheet, but it is just not pulling in the numbers from there. What could be wrong?
BTW, I saw your book on the Cochin airport book stall !
Rgds,
Vandhana
1. If we are following the 200 DMA based approach as against SIP, then we would invest cash accumulated till then as soon as Nifty goes below 200 DMA and not every day that Nifty is below 200 DMA right? In this post what is the assumption made?
--> That is correct. In addition I am assuming a small amt would be invested each day the nifty is below 200 dma
2. Also, if market timing is the objective, then every time Nifty goes above 200 DMA, then you would also redeem and invest back only when it goes above 200 DMA and your exit load period is over. While in practise this may be cumbersome, (not to mention transaction costs) mathematically, how does that pan out?
--> I have not tried this, but a similar study is available here: Is it possible to time the market?
3. Is there a base excel sheet (Like your other calculators) from which these graphs are derived?
--> Yes, but not in a format that anyone can use like the other sheets 🙂
On a different note, much as I try I am unable to use the direct equity calculator that you had posted some time back. I did log in into value research and then try the sheet, but it is just not pulling in the numbers from there. What could be wrong?
---> That does not need VR data only morningstar or the recent on screener.in
BTW, I saw your book on the Cochin airport book stall !
--> Good to know. Thank you.
Dear Pattu Sir,
Thank you for the detailed reply. Let me try the latest version of the equity calculator then.
Rgds,
Vandhana