DCA and Increasing this with the dip.

[TBS]

I will commit to making this investment when the Shiller S&P P/E falls to a more reasonable level...
https://www.multpl.com/shiller-pe

You do realize that the Shiller P/E ratio cannot be used to predict short-term market movements? And that markets can still rise even while the Shiller P/E ratio is falling?

Before we lose this thread into further walls of text, a few of the earlier points are worth repeating:

- Even if DCA outperforms lump-sum investing over the next 20 months, it cannot be said from this that DCA is the better strategy in general.
- The framing of the experiment isn't actually what is relevant to individuals here. Each individual should consider the date when the cash first became available to them to invest as the baseline, rather than 2022/02/05.
- DCA is fine as a tool to help people manage their emotions in the event of a crash, even if the expected payoff is lower compared to lump sum investing.

It is heartening to see that majority of contributors on this thread are in general agreement on this. And that people holding cash are mainly doing so out of circumstance, rather than confidence in an innate ability to predict short-term market movements

BTW @Tkydon all your quoted figures for the eMAXIS Slim All Country example are wrong - you've used an incorrect fund price for 2022/03/05.

[Tkydon]

Thanks for catching that. In my haste I mixed up eMAXIS Slim全世界株式(3地域均等型) and eMAXIS Slim全世界株式(オール･カントリー)
I went back and corrected it.

[banders]

https://ofdollarsanddata.com/dollar-cos ... -lump-sum/ is the usual position on the difference between DCA and LSI. Statistically, two thirds of the time LSI wins. So the obvious deduction would be LSI is “better statistically”. But I think this becomes a somewhat spurious/redundant statistic if you consider the current situation. At any time, not just now. If I read that Valentine’s Day has been dry for 80 of the last 100 years and I look up on that day and see black clouds, that statistic has no meaning today. No one can predict the future but you also can’t ignore the current state of the market. If we were to drop randomly into the stock market at any time in the last 100 years, statistically we would have a 66% chance of winning with LSI. But we aren’t dropping randomly into the market. If you accept the market is overpriced - and there seems to be general agreement on that - then doesn’t that long-term statistic become somewhat irrelevant?

I’m DCAing a lump sum over the next two years because that statistic doesn’t mean much to me right now. I may be end up being wrong, of course, but it’s a decision I have to make.

[adamu]

Statistically, two thirds of the time LSI wins. So the obvious deduction would be LSI is “better statistically”. But I think this becomes a somewhat spurious/redundant statistic if you consider the current situation.

Exactly. Probability being on your side doesn't mean you will win. There is not much in it and sitting on a lump sum out of fear is worse, so the rational thing to do is to lump sum, but if you are not comfortable with that, DCA is the next best choice.

[TBS]

Statistically, two thirds of the time LSI wins. So the obvious deduction would be LSI is “better statistically”. But I think this becomes a somewhat spurious/redundant statistic if you consider the current situation. At any time, not just now. If I read that Valentine’s Day has been dry for 80 of the last 100 years and I look up on that day and see black clouds, that statistic has no meaning today. No one can predict the future but you also can’t ignore the current state of the market. If we were to drop randomly into the stock market at any time in the last 100 years, statistically we would have a 66% chance of winning with LSI. But we aren’t dropping randomly into the market. If you accept the market is overpriced - and there seems to be general agreement on that - then doesn’t that long-term statistic become somewhat irrelevant?

Then instead of the average 66% statistic, look at statistics for when things looked objectively similar in the past (black cloud days, noisy forums & media about overpriced markets...). I did this for the Shiller P/E conditions @tkydon spoke about and found lump sum investing still won out more times than not against DCA. The article you linked to does something similar, yet you disagree with it's conclusions

BTW I don't accept the market is overpriced - in fact I have no idea whether it is overpriced or not. I have no idea what the market will do over the short-term.

Things I am confident about: it will go up over the long term, and more people lose out than succeed by trying to time the market

[Haystack]

If you are facing decision paralysis, split the difference.

1/2 Lumpsum 1/2 DCA.

[banders]

Then instead of the average 66% statistic, look at statistics for when things looked objectively similar in the past (black cloud days, noisy forums & media about overpriced markets...). I did this for the Shiller P/E conditions @tkydon spoke about and found lump sum investing still won out more times than not against DCA. The article you linked to does something similar, yet you disagree with it's conclusions

BTW I don't accept the market is overpriced - in fact I have no idea whether it is overpriced or not. I have no idea what the market will do over the short-term.

Things I am confident about: it will go up over the long term, and more people lose out than succeed by trying to time the market

I didn’t disagree with its conclusion; which is that LSI wins (or ‘has won’, I should say) 66% of the time. I just think the market is ready for a correction and that changes the probability.

And if you place so much faith in past statistics, why do you think the Shiller no longer indicates a correction is due, as it always has?

[TBS]

I didn’t disagree with its conclusion; which is that LSI wins (or ‘has won’, I should say) 66% of the time. I just think the market is ready for a correction and that changes the probability.

Respectfully, you did disagree with the article. There is a section on "What About Valuations?". It discusses how valuations change the probability but it still recommends LSI over DCA at times of high valuations.

And if you place so much faith in past statistics, why do you think the Shiller no longer indicates a correction is due, as it always has?

The Shiller PE is useful for predicting returns in the medium to long term (5+ years), but not in the short term. Historically it is only weakly correlated with short-term returns, and this correlation is not enough to indicate that switching strategy from LSI to DCA is justified.

[Bubblegun]

Well I have learnt something this week after my little experiment.

As you remember I decided to increase my payments into my NISA, over the past few months since January.

First thing I learnt was to IGNORE the noise from the news.

Second thing I learnt was, it was worth it and don't bother trying to time anything. (don't be greedy, and end up loosing, be satisfied with what we have). It was so tempting to try and wait and wait for a little crumb extra.

Third thing was, If there is ever that huge crash, I won't panic, I'll ignore the noise and keep dropping the money in. ( if I can).

I could have dropped all the money in, in one go, but decided to drop the money in every few days/weeks. I think in this case, it worked out better dropping in the money over this shorter period as it was excess money, It's not like millions of yen that some people have but a years worth is OK.

[banders]

Respectfully, you did disagree with the article. There is a section on "What About Valuations?". It discusses how valuations change the probability but it still recommends LSI over DCA at times of high valuations.

I read that part on valuations but misunderstood. I thought that chart was saying that when the CAPE was >75th percentile, DCA underperformed 5% of the time. It’s actually saying DCA underperformed by an average of 5%. Thank you, I stand corrected. I’m reappraising my approach