Dollar cost averaging ETF's with a CAPE ratio twist

Discussion in 'Share Investing Strategies, Theories & Education' started by rizzle, 14th Apr, 2020.

Join Australia's most dynamic and respected property investment community
Tags:
  1. rizzle

    rizzle Well-Known Member

    Joined:
    18th Jun, 2015
    Posts:
    431
    Location:
    Melbourne
    I'm toying around with my long term ETF dollar cost average buying system.

    It goes like this:


    Use the Shiller CAPE ratio as a guide to how much to invest:

    1. When the CAPE ratio is 1.5 standard deviation (sd) above long term mean CAPE ratio (SP500), then invest 70% of usual monthly allotments, saving the remainder in cash or offset
    2. When the CAPE ratio drops back below 1.5 sd, invest 130% of usual monthly allotments until 'excess' from scenario 1 depletes
    3. At all other times, invest the usual allotment amount
    Standard deviation cutoff and reduce spend amounts are both arbitrary numbers (I wanted the 'caution' trigger happening enough to be useful, but not too often to be impractical). Using the above system, with CAPE ratios going back to 1881, the reduced spending trigger occurs in 7% of all months.

    Charted, it looks like the below since 1900 (the labels don't show it, but data goes to April 2020):

    upload_2020-4-14_8-53-27.png
    Note: orange = reduce spend triggered

    I'm looking to systemise my 'market timing' approach, to avoid letting my emotion or judgement get in the way. Now my question is, does anyone know how I can simulate the returns of using this strategy vs. the benchmark (i.e. just dollar cost average the same amount every month, with no CAPE adjustments)?

    I want to know if it is actually worth the effort in overall returns (I'm guessing not but it's been a fun exercise to do anyway).
     
    S1mon, bamp, mrdobalina and 1 other person like this.
  2. rizzle

    rizzle Well-Known Member

    Joined:
    18th Jun, 2015
    Posts:
    431
    Location:
    Melbourne
    A better chart with CAPE standard deviation from long run mean on the x axis (and 1.5sd 'trigger' line in orange). So largely this strategy would only adjust the DCA averaging at market extremes (7% of all months).

    upload_2020-4-17_7-55-18.png
     
  3. mtat

    mtat Well-Known Member

    Joined:
    7th Sep, 2019
    Posts:
    328
    Location:
    Sydney
  4. rizzle

    rizzle Well-Known Member

    Joined:
    18th Jun, 2015
    Posts:
    431
    Location:
    Melbourne
    George Smiley and Terry_w like this.
  5. Sticky

    Sticky Well-Known Member

    Joined:
    2nd Jul, 2015
    Posts:
    73
    Location:
    Melbourne
    Here is an example of normal dollar cost averaging in action.
    Let's say you started investing at the worst time - peak of the market.
    You purchase the same amount every week at Friday's close price.
    Chart below shows the percentage gain on each purchase.
    upload_2020-5-28_12-6-31.png

    With DCA, you don't need a full market recovery to be ahead. Even without a twist, the math works to your advantage.
     
    Never giveup and Terry_w like this.
  6. mtat

    mtat Well-Known Member

    Joined:
    7th Sep, 2019
    Posts:
    328
    Location:
    Sydney
    Yeah but I want that super predictable 19.49% short term gain and if the market won't give it to me I will be angry.