In this third and final part on "how to select an equity mutual fund", we will consider different ways in which one 'make a choice' from a shortlist.
Yes, there are several ways to choose from a shortlist, just like there are several ways to make a shortlist. You will have to take your pick. It is easy to do so. All you need to do is to try and sell your pick to your rational side. That is you must be able to justify your method to yourself. Forget about what others think.
Before we proceed, a suggestion to read the first two stages in the selection process, if you have not done so:
(1) The objective or the preamble for fund selection
Note: The title of this post is 'making a choice' and not finding the best fund from the shortlist. This means that no matter how hard you try to shorten your shortlist, you will not be able to do so (if you objectively go about it).
So from a 10-15 fund shortlist, you will end with maybe a 2-5 fund shortlist. You will have to pick one from this. There is no other way of doing this.
This is a long post. A total of 7 methods are listed. Not all methods are (or can be) discussed in detail. Please consult the links provided for more information.
Method 1: Just pick one from the shortlist!
Method 2: Using short-term risk-return metrics
Method 3: Using long-term risk-return metrics
A. With Morningstar India
B. With the freefincal risk-return analyzer
Method 4: Using downside capture ratios
A. With Morningstar India
B. With the freefincal risk-return analyzer
Method 5: With rolling returns
Method 6: With information ratio
Method 7: With Ulcer index
I urge you read/consider all the methods, and develop your own. A combination of methods can also be considered.
Method 1: Just pick one from the shortlist!
By the time you are finished with this post, chances are that you might agree that doing an "Eeny, meeny, miny, moe" from the shortlist is a pretty good method. Why not?
Think about it, you have made a shortlist based on consistent performers based on 3, 5 and 10-year returns. So why not pick one from this list? The risk associated with the returns obtained will be neglected, but this method is any day better than what many people do: ask for best fund names or ask other people to pick from their shortlist.
If the following appears a bit complicated to you, adopt method 1, but do not fail to learn "how to review your mutual fund portfolio".
If so, why can't I use star ratings as a guide? From my experience, people who start using star ratings do not know when to stop. They would compare peers without checking the actual performance of their portfolio. I can never feel comfortable with such a strategy. Personal finance is an individual race.
Not convinced? Ty this: The Trouble With Mutual Fund Star Ratings
Method 2: Using short-term risk-return metrics
This is the method outlined in the step-by-step fund selection guide
There are two issues that I did not realize/know when I wrote the above.
A. The risk-return metrics listed at Value research are over only 3 or 5 years only. They have not bothered to state this, nor replied to my question reg. this. My guess after hours of staring at their listing pages is that the metrics are over the last 3 years.
B. See below (perhaps not immediately!)
When we make a shortlist considering 3, 5 and 10-year returns, should we only consider 3 years risk-return metrics?
Perhaps not. It is not a terrible choice, but I think we can do better.
For the record, let us get through with it. It is anyway a good starting point for getting comfortable about these metrics. To be frank, I am delighted that many readers have done exactly that.
Let us look at some definitions first:
Alpha It is a measure how much it has outperformed its benchmark index. Higher (positive!) the alpha the better. Beta Measure of volatility. Beta of 1.3 means fund is 30% more volatile than the market as a whole. So low beta is important for
Beta Measure of volatility. Beta of 1.3 means fund is 30% more volatile than the market as a whole. So low beta is important for risk-averse investors
Standard Deviation: Higher the std. dev. higher the fluctuations in returns.
Sharpe Ratio: Measure of return wrt to risk. A good fund gets returns without too much risk and will have higher Sharpe ratio A risk-averse investor will look for reasonable returns and with low fluctuations in return. So they will look for reasonably high alpha, low beta, low R-squared, low std.
Sortino ratio: The Sharpe ratio considers both positive and negative excess returns (wrt risk-free rate). The Sortino ratio considers only the negative excess returns while calculating the standard deviation. Higher the better.
If you are looking for an easy-to-understand visual aid for understanding these metrics, try this: Visualizing Mutual Fund Volatility Measures
What now?
Look for funds with reasonably high alpha, low beta, low std- dev, high Sharpe ratio and high Sortino ratio.
Hang on.
Higher the beta, higher the standard deviation (typically!)
Higher the alpha, higher the Sharpe ratio (typically!!)
Higher the Sharpe ratio, higher the Sortino ratio (typically!!!)
So all we need to do is to look for fund wth reasonably
high alpha, and low beta
I would prefer to state this the other way:
1) Look for low beta
2) choose funds with highest alpha among the low-beta group.
Pick one of those funds. You done.
I have removed a measure called R-squared (tells you how closely performance is related to the index. Higher the value closer will the performance. All large cap funds will have high R-square). I feel that this is too useful.
So let us now return to the shortlist.
Recall the procedure for getting the above return screen at Value Research. You will find the steps here
No need to do that right now. Follow this link in another browser tab to get this screen.
https://www.valueresearchonline.com/funds/fundSelector/default.asp?cat=8&exc=susp%2Cdir%2Cclose%2CnotRated
You have already made the shortlist (funds corresponding to blue rectangles)
Now click on Risk-stats
You would need to prepare this kind of table. You can do so manually. Over the years, many readers have sent me their excel files in which they have made this process simpler.
You can download the excel file from VR and extract data for the funds in your shortlist.
If you know a bit of excel, you can select the tabulated shortlist, go to data and click on filiter. That will produce the encircled sorting menu as shown above. If you don't know how to do this or use Excel, do it with pen and paper. No big deal!
You can then sort the beta column from smallest to largest. What has been described above will take only 10 mins.
Now notice the yellow selection. Low beta with reasonably high alpha.
We have our winners! Pick one fund you are comfortable with. Done!
Wait a minute! What about Franklin Blue Chip and HDFC Top 200! Why did they not make the cut?
Simply because they have not done well in the last three years and we are only looking at last three-year metrics!
Method 3: Using long-term risk-return metrics
What if you wanted to consider risk-return metrics over longer durations? After all, we considered long-term returns to make the short-list.
Makes sense. There are two ways to accomplish.
A. With Morningstar India
There are two issues here. (1) there is no risk-return listing for all funds like VRonline and (2) all ratios are calculated with BSE 100. Not a bid
(1) there is no risk-return listing for all funds like VRonline. So you need to go to each fund page.
(2) all ratios are calculated with BSE 100. Perhaps not a problem for pure large cap funds. Will not use for other categories.
See here for a sample. This is not a screenshot. I have only included stuff that I need!
Notice that the metrics are available for different durations.
Trouble is, the funds benchmark is BSE 200 and not BSE 100.
You can consider the metrics pretty much as mentioned before, make a table for different durations or visually observe for consistency and make a choice. It will not take as long as it seems. Hey, whose money is it anyway!
This has a got a new metric:
Treynor Ratio is known as the reward to volatility ratio. While the Sharpe ratio is the excess return (wrt risk free rate) divided by standard deviation, Treynor ratio is the excess return divided by beta. This is calculated for both the fund and the benchmark for which beta is assumed to be 1.
Higher the Treynor ratio, better is the performance (higher returns + low volatility wrt benchmark)
B. With the freefincal risk-return analyzer
If you do not like the fact that morningstar uses the same benchmark for all funds, you can consider my risk-return analyzer
You can then choose from Nifty, CNX 100, CNX Mid Cap, CNX 500, Sensex, BSE Small Cap, Mid Cap, BSE 100,200, 50.
Consider 1 to 9-year durations. Calculate lump sum and SIP returns for each duration
Evaluate the risk-return score with 13 metrics.
You can also consider the Year on Year Mutual Fund Risk Return Analyzer for reviewing yearly performance.
Method 4: Using downside capture ratios
This is the method I advocate in the investor workshops.
The main advantage of this model is that it does not depend on modern portfolio theory (MPT) ratios which are applicable only for normal or Gaussian distributions. It is quite possible (in fact easy to pve)that they may not work with mutual funds and stocks! This means all star ratings are on shaky ground!!
People use MPT metrics because alternatives are tough to evaluate (more on this soon).
Upside capture.
For a given period, how much of the benchmarks gains has the fund captured? Higher the better.
Downside capture.
For a given period, how much of the benchmarks losses has the fund captured? Lower the better.
From what I have seen, consistent downside protection is the source of alpha.
I have written about this in detail earlier:
Understanding Upside and Downside Capture ratios
Simplify Mutual Fund Analysis with Upside/Downside Capture Ratios
Again two methods
A. With Morningstar India
Morningstar again uses only BSE 100 for calculation. However, this is easy to use and is perfect for large cap funds and maybe a bit too stict for mid and small-cap funds.
B. With the freefincal risk-return analyzer
Again trouble with morningstar is it uses the same benchmark for all funds. You can use my risk-return analyzer which has a separate upside downside page. The evaluation strategy is different from what morning star uses but the essence is the same.
I will post an exclusive tool to calculate this better. See some screenshots of this are available here
Method 5: With freefincal rolling returns calculator
Ro0lling returns offer a smart and easy to understand way to evaluate consistency in performance.
Suppose you want to evaluate fund performance from Jan 1st 2000 onwards.
You calculate 3-year CAGR* from 1s Jan 200 to Dec. 31st 2002. Then you shift the period by a day (roll over) and calculate 3-year CAGR* from 2ns Jan 200 to 1st Jan. 2003 and so on. Plot them together and stare at it.
- for both fund and its benchmark
This idea is to look for consistent long-term outperformance.
See this post for more details:
Mutual Fund Rolling Returns Analysis: Franklin India Blue Chip Fund
and use this automated sheet for calculations:
Mutual Fund Rolling Returns Calculator
Method 6: With information ratio
Information is defined as the average excess return of the fund wrt benchmark for a given period, divided by the standard deviation of the excess return.
Consistently high information ratio is a good indicator of performance. Dr. Uma Shashikant says,
“Information ratio is my key quantitative indicator”.
See this post for more details (calculator in the same post):
Mutual Fund Analysis with the Information Ratio
Method 7: With Ulcer index
Ulcer index is another downside protection indicator. It penalizes funds more for lapses in downside protection than other metrics. This is my personal favourite (thanks To Ramesh Mangal)
See this post for more details:
Mutual Fund Analysis With the Ulcer Index
and use this for calculations:
risk-return analyzer which has a separate ulcer index page.
Was I not right? Doesn't "Eeny, meeny, miny, moe" from the shortlist seem like a pretty good method now!!
Would you like a suggestion?
For large-cap funds:
Use: Method 4 A : Using downside capture ratios with morning star data
For mid and small-cap funds
Use: Method 4 B : Using the new downside capture ratio calculator (releasing in a couple of days)
Alternatively
Use: Method 7: With Ulcer index
or a combination of
Method 5: With freefincal rolling returns calculator and
Method 7: With Ulcer index
Part 1: How to select an equity mutual fund -preamble
Part 2: How to select an equity mutual fund – Creating a shortlist
Download the combined PDF version of the updated mutual fund selection guide
Register for the Hyderabad DIY Investor Workshop Nov 27th 2016
Check out the latest mutual fund returns listing
Buy our New Book!You Can Be Rich With Goal-based Investing A book by P V Subramanyam (subramoney.com) & M Pattabiraman. Hard bound. Price: Rs. 399/- (Rs. 359/- at Amazon.in). Read more about the book and pre-order now! |
I don't think any financial advisor in India puts this much effort to pick and choose the right mutual fund scheme for their clients. Before even making any shortlist they mostly follow "Eeny, meeny, miny, moe” method. Though I have doubts how many investors will actually be willing to follow the simple steps mentioned by you in your last 3 posts. Not many. But you are doing your bit and more. SEBI should thank you and acknowledge your effort.
Thanks Sandip. Over the years, many investors have used the method I had outlined in the earlier version (using short-term risk-return metrics). That is many, relative to what I had expected. Agree that the majority of investors as a whole will not be willing to follow these steps.
I really appreciate the effort you spent on long hours to come out of this selection process. Even i have seen the Financial planners are taking very little time in selecting the funds for customer , purely based on last 3,5 years performance provided by VR online.
Some of them even not suggesting QLTE, because they are not getting commission from Quantum AMC.
But you are helping lot for retail investors to understand the process of selection with out expecting any benefits. Hats off you Mr.Pattu.
Thank you very much.
Add me sir
Method 1 is what I follow. Even for coming out of a fund, I come out when it is no longer in the short list (but giving some benefit of doubt for barely missing). For e.g., I invest in 1 large&mid cap and 2 mid & small caps funds. The large cap funds I have in my portfolio are hdfc equity and icici dynamic. Both are still good funds but no longer make my short list. So coming out of them and will be out in the next one year and routing the redemptions and fresh sips to franklin prima plus. And will check the present fund's position every year or two. The 2 mid & smallcap funds still make my short list and are so no portfolio rearrangement.
My reason for not using risk parameters is I believe even they are a sample of a random process. I did not do the analysis but if we check at the second order statistics of risk parameters, I believe they also will have good variance. If I am right, we can add another caveat saying "Past risk is not indicative of future risk :-)"
Every method has its disadvantages. Risk metrics are no different. Imo they provide an insight into volatility when used right.
What I meant was any of these methods is as good as any other. May be we can do a study on how each method would have performed and see if there is any significance in a good confidence interval between the results of each. How easy it will be for you to do such exercise? I would assume near impossible due to dependency on VRO. But if you have required data not time, I would be glad to do such exercise. While I believe there will not be a clear winner in the statistical sense it would be good to know to choose better method in future.
I have done such an exercise and there is not much difference in returns. That is, however not the point. Neither is finding a clear winner. My idea is to choose a fund which has a good history of return AND a good history of downside protection (which can be measured in various ways).
Thanks a lot. Do you have a blog post of about that analysis? About finding a clear winner, I believe is impossible. One may be lucky once or twice but there will be no algorithm that can reliably do that.
I think as you said in other posts investing in mutual funds is very simple and easy. It is even easier that what we make out it to be. It is more about patience and participating. Method 1 working as well as others is a good thing in that way 🙂
Also thanks for the work you did on this guide.
My sincere thanks to you for your time and effort in showing a road map to fund selection.
Being a starter in the field when trying to struggle to shortlist from fund universe your blog lightened me up.
With your detailed step by step analysis, now, I can prove to my managers that I got more knowledge and can do more systematic fund selection.
None of the online posts does give such detailed step by step approach, then finally wondering whether to buy a book or do CFA!!!!
I’m very happy to convey my best wishes to you and keep posting such valuable posts at your free time.
Thank you very much
O(f(our Thank yous) /g(Your Work)) --> inf
🙂 Thanks man.
OMG this a highlight post Puttu sir! I can imagine the amount homework gone into this post. Appreciate your effort.
Eeny meeny mine moe,
pattu sirru kku poodu ooo!
do you really have any work teaching at IIT, is it that easy ?? hmmm... 😉
Thank you. Teachers get vacation!
Oh! you are using your vacation for this....
........that is passion for being generous.
its hard to find people with this mindset! Thank god for creating these good souls!
Regardless of how much you research you will get either HDFC top 200 or Franklin India Blue Chip
Nope. Not true.
Hi Mr.Pattu ,
Can we use the same Method 4 A : Using downside capture ratios with morning star data for selecting Large + Midcap or Diversified equity funds.
If possible can you share the output of all these selection final list of funds.
The final list will be useful to ensure whether we are selecting as per the guide. Or we miss some key parameters in selection.
Hi Pattu,
Thanks again for consolidating all related to MF investing in one. Very useful.
Regards,
Use: Method 4 B : Using the new downside capture ratio calculator (releasing in a couple of days)
Are you still preparing the new calculator for Method 4B.
Eager to use that new downside calc for the mid/small cap selection .
Yes. Will do so this week.
I have gone through the fund selection.I have understood that lower the beta higher the alpha.Can you help me in Standard deviation, Sharpe ratio and Sortino ratio and Rsquare.
How do we prioritise like for Example: 1. Lower beta 2. Higher alpha 3 low standard deviation
that is in other way-like a rule-BODMAS rule in maths
Sir, What do Negative downside capture ratios mean. For Eg -->
http://www.morningstar.in/mutualfunds/f0gbr06r75/sbi-magnum-global-fund-growth/risk-ratings.aspx
(Under the 1-Year Category)
Hello,
I am MF investor since 3-4 years are read very often several blogs from some renowned portals, today eventually land up on your blog..and got simply mesmerised and astonished by reading....
the main things is... the amount of explicit information and tools you are making available for readers is simply exemplary and amicable... i am sure other financial expert do have this knowledge, what i mostly realise its mostly very generic.... dont know why they dont share (may be for very simple reason... 🙂 .....
i have now subscribed to your blog....
I wish you lot of success in this endevaour....
Regards
Amit
Pattu - you are killing the MF Advisory services business 🙂 The value of your posts in helping a DIY investor is just invaluable! I took it over from my commission based advisor totally!
Thank you 🙂