Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion

Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.

    Support MFO

  • Donate through PayPal

How to Calculate Risk-Adjusted Rate of Return

edited January 2013 in Off-Topic
"... if a particular investment class is on a roll and does not experience a great deal of volatility, a good return per unit of risk does not necessarily reflect management genius. When the overall momentum of technology stocks drove returns straight up in 1999, Sharpe ratios climbed with them, and did not reflect any of the sector’s volatility that was to erupt in late 2000."


  • edited January 2013
    Morn'in hank,
    Only one cup of coffee here, so far; but I'll try to be somewhat alert. Tried an early morning walk about; but it is too damn cold outside....but sure was a good face slap to my physical body, to wake up.
    You posted questions from the "burning leaves" thread, which relate to your note here and the link. Thank you for the link. This subject was going to be part of a Funds Boat write; but this thread serves that purpose.
    "Catch: You've kept no secret of your returns. But, I'd love to hear you quantify (1) the yardsticks you use to measure the risk of various assets you own and (2) the tools you so employ to obtain these measurements. I'd expect that to be a daunting task - as there seems little agreement here or in the financial media regarding varying degrees of "risk" re: cash, bond, commodity, real estate, and equity sectors."
    The yardsticks tend to be more personal for our house versus traditional investment yardsticks/benchmarks for many funds, at least related to active managed bond funds.
    On the personal side, is that we literally attempt to average an 8% annualized return over any 5 year period, regardless of where the gains come from. Other personal areas are stoppage of cash flow from work sources and the related ongoing expenses that affect our own personal budget. We can not "makeup" losses from a bad market period via continued dollar cost averaged inflows into a 401k or similar account; as that avenue is now gone. Budget related, aside from the normal areas of property taxes, food, gas, utilities and etc.; is healthcare. Relative to retired boomers in MI, we are in another class in the healthcare budget. From my non-scientific research, meaning folks/family I personally know, is that about 70% of those retired have healthcare benefits, that our house does not have. We both had enjoyable work environments in our chosen fields, but neither involved benefits of post-retirement connected to being members of a union. The union folks/family we know are mostly retired General Motors, teachers and a few skilled trades union retirees. A hardy discussion (12 people) about all of this happened to take place this past holiday period; and included some younger folks (early 40's), too; who were able to discover real world conditions. We were the only two who did not have union connected jobs. Although post-retirement healthcare had been discussed, now and then, over the past 10 years; it appeared that the majority of the issues had been forgotten by others. Post-retirement, Medicare and related cost about $4,000/year for those without any post-retirement/union benefits backing and coverage of other medical/dental costs; in order to be properly covered and protected around the deductibles and holes in Medicare and to have a dental plan in place. While $4,000/year/per person may not be much money to some folks, it is still real money that we must spend and account for going forward in one of the most inflationary sectors around. Kick this out for 20 years, modified with inflation; and it is real money, eh? All of this related to the home budget is the personal risk factors for investments.

    As to risk factors related directly to investment sectors; this task is less clear for the many reasons that have discussed here........70% market volume trading by machines, perverted money policies by central banks and at least for the U.S., and other developed countries, an industrial era that in no way resembles what you and I are aware of from the past several decades; and will never be again.

    A risk benchmark example:
    I find little value in current bond fund benchmarks that use AGG; except to discover whether a bond fund may actually be ahead of AGG. This is no surprise and is part of the continued expansion of "modifed" active managed mutual funds and relationships to some benchmarks used by M* and others. This is not a failing of M* or others; as how in the heck does one benchmark a PONDX to something relative to this bond fund. Nope, can not do this. We, our house, is not with any illusion as to how PONDX was able to return 20% in 2012, while AGG could only provide 4%.
    Most of us with active managed funds are within two risk factors, regardless of the market area one's monies are invested. We are at risk of a particular sector not doing well for whatever reason; and also at risk with the abilities of fund management missing the cues of market moves and not using the "tools" allowed within the fund properly or poorly, or both.
    The best any of us may do to establish our own risk, is both of a personal nature; and an understanding of whether one's active managed funds are more plain jane or on "steroids".
    If our household was sitting upon $5 million of disposable/investable income; although we wouldn't like to lose 10% or more of the invested monies in a year, we could survive, eh?
    Obviously, all of this is the logic behind any one investor's investing decisions.
    "I'd think both the yardsticks used to quantify risk as well as the specific techniques and instruments employed to measure it vary greatly in the real world. John Hussman calls HSGFX "risk adjusted" - but it turned out a negative 12% last year. I'm sure in his mind that was the price that needed to be paid in the interest of reducing risk for his investors. I also suspect David Tice or Jim Rogers. uses a different risk metric than say - the folks at Dodge & Cox or T. Rowe Price. So ... let's hear yours. The returns you've portrayed. What makes you believe they were praiseworthy on the risk-reward spectrum? In other words, how do you know they were achieved with relatively little risk?"
    I don't and won't say that any of our returns were without risk or low risk. Without or low risk is only a function of hindsight; discovering one was in the right place at the right time, and thus the risk was okay. Real risk for our house, and what is likely part of most decisions for all investors is how and where to invest going forward. I could easily state that what we have used for cash holdings, which is primarily FINPX, ACITX and PLDDX, performed well over the past few years; and they have. We're just not a traditional cash position house; with the exception of mid-2008 through May, 2009; so some money may be parked in these areas are various times.
    We don't find anything less challenging about bonds than any other sector of investments. Some of our portfolio is at risk right now; depending on how far and how long traders want to push and play with yields. Technically, we are already down 1.5% for the year; had we been invested in a 50/50 equity and bond mix. Will the big equity moves in this early stage of the year hold? I doubt it. Some trading houses are going to gather the equity profits; and some others will move back into some investment grade bond sectors, including government issues.

    You asked:
    "The returns you've portrayed. What makes you believe they were praiseworthy on the risk-reward spectrum? In other words, how do you know they were achieved with relatively little risk?"
    I don't know that our 2012 returns were praiseworthy; but we are satisfied and "got lucky", eh? Hopefully, what was written above helps define the risk factors for us; both personal and related more directly to a particular investment.
    Mr. Hussman and others will all have their comfort zone based upon their knowledge and methods. Although I surely don't understand Mr. Hussman's methods; PhD or not. Mr. Rogers; well he can afford to make a large mistake, that our house can not.

    Our portfolio may indeed be on the edge of adjustment. I noted in one report (don't recall which date) about the flattening of some bond funds returns during the past few months.
    I will note our "looking at" equity list, when I have time to post the Funds Boat this weekend; but has some choices as noted in a reply to "forch" in his portfolio building thread from a few days ago.
    Lastly, as to equity or bond portfolios; remains the question of to go with "steroid" active managed funds or a more straight forward etf or index.
    As to the subject line of the post, How to Calculate Risk-Adjusted Rate of Return; one may be able to get a feel for risk via a particular investment sector, but unable to properly measure an active managed fund. Risk adjusted returns may be an easierly task with a more pure etf or index.

    The preferred method of chat for this would be the 3 cups of coffee meeting, at the local restaurant, eh? Not so easy to find and/or place all of the words needed for a proper answer.

    Take care, up there,
  • beebee
    edited January 2013
    Hey Catch and Hank,

    I also believe risk is reduced after market corrections. Just as markets get more and more risky (in a bubbled up sector) it also becomes less and less risky as those assets are repriced due to a severe correction. I just bought a condo for $35K that my neighbor paid $165k for prior to the real estate bubble. I would say his timing and my timing will be the determining factor of real return on our two very equivalent assets. Good fund managers try to find these opportunities.

    In PONDX case, (if I understand it correctly) many of its bond securities were purchased at considerable discount to par. One coud argue that credit risk is a larger issue with these securities so no free lunch. PONDX has a management team that has invested in deeply discounted securities that, not only pay a dividend, but also have the opportunity for capital appreciation. This type of bond manager removes considerable downside risk it its holdings by finding these deeply discounted assets much like a deep value stock manager finds stocks selling at a discount. I believe there are many more of these values still to be found... whether it is a bond or a stock or a condo.

    Appreciating assets often don't just stop appreciating...they reverse direction...quickly. Keeping these investments on a short leash and gleening periodic profits can be ways to offset the growing riskiness of these assets. Taking periodic profits is a sell discipline that I try adhere to. These profits can then be DCA into other parts of your long term investment plan. As a retiree, these periodic profits could provide short term income, be used to pay down debt, or be reinvested into the next under appreciated, deeply discounted asset in your long term portfolio.

    Downturns provide the opportunity to buy things "on sale" as well as provide downside protection there by reducing risk and improving total returns. Selling appreciated assets is a hard discipline to adhere to but just as important as DCA into an investment. I think we need new acronyms for these two investment strategies:

    DCAin = Dollar Cost Average into an investment
    DCAout = Dollar Cost Average out of an investment

    Any seeing buying opportunities out there?
  • Hi Hank. I remember there was some discussion around the Sortino ratio a couple years ago on this board. I liked it's premise that it measured down-side risk and could help point out fund managers who handled downside risk better then others. I remember searching for this ratio but I never did find a place where you could get the information free. I believe BobC. said he had purchased software to get these values.

    The newish M* upside/downside capture ratio they give in there statistics is what I've come to like quite a bit in picking less risky funds. For example, if a given fund captures 98% of market during good times but only looses 76% in bad, that would be a pretty good fund over a economic cycle. Those values actually are the upside downside values for YAFFX over the last 10 years.

    I used to do my own risk analysis by just calculating the probability of what a fund could make or loose in any given year. Probability is just a calculation using the funds average returns and it's standard deviation. Using 1x stdev would give you a probability confidence value of 68%. 2x stdev would be 95%. So for example: YAFFX over 10 years has an avg yearly return of about 11% and a stdev of about 16%. That would say you can be 68% confident in any future year YAFFX will return between -5 and +27%. Of course that also means 32% of the time the returns will fall outside that range. Taking 2sigma (2x standard deviations) would give you more confidence in the return range, 95% confidence. Then the expected return range for YAFFX would be -23 to +43. The 95% confidence number for the low end of the range happens to be pretty close to YAFFX's worst year in 2008. In 2008 YAFFX actually lost -23.5%. Statistics, go figure.

    Thanks for the article.

  • edited January 2013
    Interesting discussion:

    Another link for risk adjusted returns is here:

    All 3 measures can be used for comparison purposes. Although they are discussed under risk adjusted returns, in my opinion, they are not really returns but risk measures derived from returns.

    Personally, calculating these measure for ones own is somewhat problematic as most people do not have daily (or even weekly) return information and going back from the last portfolio composition is problematic as portfolio changes during the time frame. But it is doable. Assuming the portfolio has not changed much, measures involving weighted beta (e.g. traynor ratio) is probably easier to evaluate based on final portfolio composition alone.

    Other heuristic measures are possible as well.

    For example: compute Ra = Rp * (Bb/ Bp) then compare Ra with Rb


    Ra = Adjusted return
    Rp = Portfolio actual return
    Rb = Benchmark Return
    Bp = Portfolio weighted beta
    Bb = Benchmark Beta

    This measure, Ra, risk scaled return, would allow you to compute and compare your adjusted return taking into consideration of beta of your portfolio and benchmark.

    You can do something similar by using Std.Dev instead of beta.

    But in the end, you can only spend actual returns and not risk adjusted returns. Neither can you eat or take shelter based on risk adjusted returns. If you need 8% return to live, it will not provide much comfort to have achieved 4% but at 1/3 the risk.

    One the return is achieved it is history and risk free. Only future is risky:)
  • Reply to @bee:
    My question to you: Is the house still there? Sounds like you bought the lot.
    Best wishes for your new investment.
  • Reply to @Derf:
    Last time I checked...
Sign In or Register to comment.