Official Shiny Things thread—Part III

[culepico]

For me, I don't care whether dip or rise. But as long as I konw for sure it will dip further, I hold for the moment. In the case of COVID, most countries still shows no sign of recovery, a very big telltale that the stock will remain or dip even lower until we start to see some recovery. Can't say if it will work in other recession case like GFC or not. But at least we have seen that fastest sign of recovery is at least 2-3 weeks after effective lockdown (South Korea). Holding DCA will also means more ammo to unload once it dips lower.

But of course nothing is 100%, if one day a drug/vaccine is announce, we will be chasing after the boat. Or if there is a burst of transmission even after the infection in stable, there will be more dips.

Beware bro, you might get flamed from the people here by trying to time the market

 

[Nesort]

Like to check, if you see this news, would you think that the dip would be over and it is actually a good time to buy now? Or should I wait a while more?

msn.com/en-sg/news/world/coronavirus-good-news-newborn-baby-beats-bug-vaccine-on-way-and-italy-cases-slow/ar-BB11mEjM

 

[kram62]

Many people have been asking for data based calculations to justify their strategies.

There is a very good blog that does just that, and it has some articles that address many of the recurring topics that are debated here.

Here is a selection of articles with data-based evaluation of various strategies that have been hotly debated recently:

- Even God Couldn’t Beat Dollar-Cost Averaging (The Problem with Buying the Dip)
- When Does Market Timing Work? (On the Necessary Conditions for Successful Market Timing)
- Why Market Timing Can Be So Appealing (On the Temptation to Have Cash on the Sidelines)

Sorry, but no TL;DR here, you need to read the articles and see the data to understand the results and conclusions.

The rest of the blog is well worth reading!

 

[moolala]

Like to check, if you see this news, would you think that the dip would be over and it is actually a good time to buy now? Or should I wait a while more?

msn.com/en-sg/news/world/coronavirus-good-news-newborn-baby-beats-bug-vaccine-on-way-and-italy-cases-slow/ar-BB11mEjM

this virus is a joke

within 1 month and it's over

 

[hkchew03]

Beware bro, you might get flamed from the people here by trying to time the market

Like it or not, this time round is not something fed pump $ can solve. The fastest way to end this recession would be a miracle drug/vaccine or hopefully the warm weather is enough to weaker or completely kill the virus. But until then, everyone including the market is at the virus mercy (even Greenland isn't spared).

If the issue is due to trade war, oil price etc, then I would continue to DCA, as it is a more logical choice.

End of the day, still depends on individual preference and risk tolerance, there is no one-only solution.

 

[Boiboi321]

Quick question -- are Singapore residents now allowed to buy Singapore stocks on IBKR?

 

[streetfighter]

https://www.afr.com/markets/debt-markets/vanguard-cranks-up-exit-costs-on-aussie-bond-fund-20200318-p54bi3

I read with great concern about the above. It means we can't sell the ETF at market price but have to sell at big hair-cut?

 

[hwckhs]

https://www.afr.com/markets/debt-markets/vanguard-cranks-up-exit-costs-on-aussie-bond-fund-20200318-p54bi3

I read with great concern about the above. It means we can't sell the ETF at market price but have to sell at big hair-cut?

I can't see anything beyond the first few lines:

You have reached an article available exclusively to subscribers

 

[streetfighter]

I held my stocks & ETFs bought since 2009 & keep accumulating until recently then i sell all, so happy i sold all to lock in the >50% profits (based on accumulated capital since i put in much more money at later stage, annualized return will be greater but i never count). So i don't have the problem you mentioned.

When to get back in? Well, i will just monitor first then decide.

If you did this last month, then you would have probably done (or in the future, will do) similar things wheneve the market goes up and is at all time highs. Between early 2009 up until last month, doing this kind of would have put a real drag on your returns.

Then there is also the question of when to get back in, most people think it will keep going lower... until it doesn’t, and even after it goes up they think it will go back down... until it doesn’t.

 

[auvignon]

Don't buy dips! Correct! That was my point in my arguments really. Read them properly again. Do you understand now? I advocate not buying dips! And I also advocate against blind DCA!

My method doesn't require timing the market at all. It is essentially just removing the component of buying down the market as compared to a regular DCA.

I think many people have a misconception of my method. It is not advocating timing the market at all. I am merely stating that by removing "DCA down the market", you will perform better than those who did it!

You can do the simulations/calculations yourself for convincing sake.

I was curious enough about culepico's methodology to do some quick calculations, although I don't have sufficient post count to post a link to said calculations.

But broadly summarising, if I set the below price parameters for ES3, and on the basis of the following assumptions: everyone starts off with nothing; the DCA person spends $100 every month regardless; culepico rolls over $100 for every month that ES3 dips, but then spends all of the rolled over amount the next month that ES3 rises (together with the $100 for that month); and, say, 2.5% dividends paid out in each of February and August), the average price of a person who DCAs every month is $3.1 whereas culepico's methodology would have resulted in an average price of ~$3.12. If we take ES3's price in December to calculate returns for this hypothetical year, the DCA person would have gotten ~6.45% (~7.12% with dividends) whereas culepico would have only gotten ~5.74% (~6.21% with dividends), a difference of ~0.71% (~0.92% with dividends).

I may be wrong (and/or my calculations flawed), but isn't DCAing through the dips meant to reduce your average price in a guaranteed fashion? How else would you consistently and predictably average down?

---

January 3.5
February 3.3
March 3
April 2.9
May 2.85
June 3.1
July 3.35
August 3
September 3.2
October 2.8
November 2.9
December 3.3

 

[moolala]

I held my stocks & ETFs bought since 2009 & keep accumulating until recently then i sell all, so happy i sold all to lock in the >50% profits (based on accumulated capital since i put in much more money at later stage, annualized return will be greater but i never count). So i don't have the problem you mentioned.

When to get back in? Well, i will just monitor first then decide.

50% over 10 years ...is not alot or is it?

 

[culepico]

I was curious enough about culepico's methodology to do some quick calculations, although I don't have sufficient post count to post a link to said calculations.

But broadly summarising, if I set the below price parameters for ES3, and on the basis of the following assumptions: everyone starts off with nothing; the DCA person spends $100 every month regardless; culepico rolls over $100 for every month that ES3 dips, but then spends all of the rolled over amount the next month that ES3 rises (together with the $100 for that month); and, say, 2.5% dividends paid out in each of February and August), the average price of a person who DCAs every month is $3.1 whereas culepico's methodology would have resulted in an average price of ~$3.12. If we take ES3's price in December to calculate returns for this hypothetical year, the DCA person would have gotten ~6.45% (~7.12% with dividends) whereas culepico would have only gotten ~5.74% (~6.21% with dividends) , a difference of ~0.71% (~0.92% with dividends).

I may be wrong (and/or my calculations flawed), but isn't DCAing through the dips meant to reduce your average price? How else would you average down?

---

January 3.5
February 3.3
March 3
April 2.9
May 2.85
June 3.1
July 3.35
August 3
September 3.2
October 2.8
November 2.9
December 3.3

You are not wrong, but you only calculated for 1 year, which is an extremely short period for DCA, and is suppressing a lot of market data don't you think? By opening up the timeline to lets say 10 or 20 years, it is a better representation of the real market data? The wider the timeline, the more realistic it is for DCA, and the data are more "unbiased".

And also for that matter, wouldn't S&P 500 or IWDA be a better representation of the world market?

You can try to calculate again

 

[Investnewbie82]

If there’s no dividend? Like IWDA..

I was curious enough about culepico's methodology to do some quick calculations, although I don't have sufficient post count to post a link to said calculations.

But broadly summarising, if I set the below price parameters for ES3, and on the basis of the following assumptions: everyone starts off with nothing; the DCA person spends $100 every month regardless; culepico rolls over $100 for every month that ES3 dips, but then spends all of the rolled over amount the next month that ES3 rises (together with the $100 for that month); and, say, 2.5% dividends paid out in each of February and August), the average price of a person who DCAs every month is $3.1 whereas culepico's methodology would have resulted in an average price of ~$3.12. If we take ES3's price in December to calculate returns for this hypothetical year, the DCA person would have gotten ~6.45% (~7.12% with dividends) whereas culepico would have only gotten ~5.74% (~6.21% with dividends) , a difference of ~0.71% (~0.92% with dividends).

I may be wrong (and/or my calculations flawed), but isn't DCAing through the dips meant to reduce your average price? How else would you average down?

---

January 3.5
February 3.3
March 3
April 2.9
May 2.85
June 3.1
July 3.35
August 3
September 3.2
October 2.8
November 2.9
December 3.3

 

[gerdhold]

The move in oil price and AUD was insane.

AUDNZD I think touched parity lol.

AUD is now emerging market currency
1y at 20 vols and skew at -7

 

[d5dude]

AUD is now emerging market currency
1y at 20 vols and skew at -7

ASX200 also crashed, this makes the country's stocks very cheap now. I had zero allocation to Australia prior to this but this crash really made me interested.

 

[culepico]

You are not wrong, but you only calculated for 1 year, which is an extremely short period for DCA, and is suppressing a lot of market data don't you think? By opening up the timeline to lets say 10 or 20 years, it is a better representation of the real market data? The wider the timeline, the more realistic it is for DCA, and the data are more "unbiased".

And also for that matter, wouldn't S&P 500 or IWDA be a better representation of the world market?

You can try to calculate again

I have done a calculation myself using S&P 500 historical data from 1 Jan 2010 to 1 Dec 2019 (120 months in total). Using the opening price at the 1st of each month as a guide, the average buying price for blind DCA was 1953.37174 while the average buying price for not buying on dips and rollover that cash to the first uptrend month was 1932.175654.

If you widen the timeline to 20 years, the difference is more. And take note, my calculation was during the bull run. If you had stretched to 20 years and factor in the 2008 recession, the difference will be more obvious.

 

[auvignon]

You are not wrong, but you only calculated for 1 year, which is an extremely short period for DCA, and is suppressing a lot of market data don't you think? By opening up the timeline to lets say 10 or 20 years, it is a better representation of the real market data? The wider the timeline, the more realistic it is for DCA, and the data are more "unbiased".

And also for that matter, wouldn't S&P 500 or IWDA be a better representation of the world market?

You can try to calculate again

Alright. I used Yahoo Finance to obtain the monthly historical price for IWDA, starting from 30 June 2012 to 1 December 2019 (no 10 year historical data unfortunately). Same assumptions as previously (i.e. everyone starts with 0 on 30 June 2012, with the first purchase on 31 July 2012; DCA person spends $100 on the first of every month, buying on the opening price of IWDA on such date (save that if the first is not a trading day, the opening price of the nearest trading day); culepico rolls over $100 for every month that IWDA dips, but then spends all of the rolled over amount the next month that IWDA rises (together with the $100 for that month); etc).

Average price of the DCA person is ~$44.77, whereas culepico's methodology will get one ~$45.22. Using the closing price of IWDA on 1 December 2019 to calculate returns, DCA person gets ~40.78% whereas culepico's methodology will get one ~39.38%.

What do you think of the results? To me (again, may be wrong), it honestly seems like the difference is too slight to spend that much effort fretting or keeping track when the whole point of adopting a passive investing strategy is not to do so.

 

[streetfighter]

Most of the money (>75%) are invested in recent 3 years, so you could not count return this way. Also i have not included div received.
Before including div, the profit is about $500k. May be this is small change to you but i am very happy with it.

50% over 10 years ...is not alot or is it?

 

[culepico]

Alright. I used Yahoo Finance to obtain the monthly historical price for IWDA, starting from 30 June 2012 to 1 December 2019 (no 10 year historical data unfortunately). Same assumptions as previously (i.e. everyone starts with 0 on 30 June 2012, with the first purchase on 31 July 2012; DCA person spends $100 on the first of every month, buying on the opening price of IWDA on such date (save that if the first is not a trading day, the opening price of the nearest trading day); culepico rolls over $100 for every month that IWDA dips, but then spends all of the rolled over amount the next month that IWDA rises (together with the $100 for that month); etc).

Average price of the DCA person is ~$44.77, whereas culepico's methodology will get one ~$45.22. Using the closing price of IWDA on 1 December 2019 to calculate returns, DCA person gets ~40.78% whereas culepico's methodology will get one ~39.38%.

What do you think of the results? To me (again, may be wrong), it honestly seems like the difference is too slight to spend that much effort fretting or keeping track when the whole point of adopting a passive investing strategy is not to do so.

I will admit in this scenario blind DCA is so slightly better. Too bad IWDA doesn't have 10 years or more worth of data .

You can have a look at mine, using 10 years of S&P 500 data. The blind DCA is slightly worse. And if you stretch the timeline even longer, the blind DCA will be more worse off.

 

[auvignon]

I will admit in this scenario blind DCA is so slightly better. Too bad IWDA doesn't have 10 years or more worth of data .

You can have a look at mine, using 10 years of S&P 500 data. The blind DCA is slightly worse. And if you stretch the timeline even longer, the blind DCA will be more worse off.

Got it. It’s interesting enough to warrant another look when time permits.