From my experience DCA will always win. The best perk of VCA is having extra cash on hand to take advantage of bear markets and underpriced stocks. Good video
I was thinking a modified approach could be the best way to go. Investing all excess money on hand with a lump sum approach but then also DCA on a monthly/biweekly/weekly approach as you earn money. If you are going to be putting money in during a down month why not really take advantage of those lower months when you can!
I was thinking a modified approach like you use could be the best way to go. If you are going to be putting money in during a down month why not really take advantage of those lower months when you can!
You should have mentioned Michael Edleson’s book, that way your audience can go straight to the font and learn about it if they want to. I do think that one advantage to VA is that one will probably spend less money out of their pocket to reach a certain goal, if that's something you want. Also, in Michael's book, VA does beat DCA if you're using quarterly investments, instead of investing monthly.
Why not mix the two options? Set monthly networth target and a minimum investment amount. Even if you are above your networth target you continue adding the minimum, but in extreme times when you are below your networth target. You add extra to keep up with the target. I think this will be riskier, but I am 23 and I am willing to take a higher risk as I have statistically many years left in my life
The conclusion of this video is wrong because it assumes a fixed growth in value each month. Thus you are comparing a linearly growing portfolio against a an exponentially growing portfolio which the DCA always produces. In that case OF COURSE you are always going to end up with a result where the exponentially growing portfolio is going to outperform the linearly growing portfolio over long enough period. This does not require any backtesting to know that result. Setting a target growth that is linear is actually a really dumb thing to do. Why do you use a fixed monthly growth in your model but not an exponential monthly growth where you aim for higher growth each next month? And then the more interesting part is what exponential power do you use to determine by how much you want to increase your target each month. Well you should use the same exponential power that a DCA approach would yield. Hence the way to PROPERLY compare the strategies is to first compute the DCA computation over the past 30 years. Then from that computation, extract the exponential power and use that to figure out where to set your value growth target for each month. Then do the Value averaging computatoin. You will find remarkably different outcome and the Value approach will always outperform the DCA. The video is totally flawed.
Im investing in a robo advisor, Betterment, that recommends a stock/bond ratio depending on time horizon. I can also auto set deposits weekly/monthly/quarterly into the taxable accounts (and they tell me the estimate of how much ill have when the time horizon comes) If I turn on their auto setting, they'll rebalance the ratio more bond less stock when i get closer to my time horizon. They also have tax loss harvesting. Would this be DCA or value averaging? It's all ETFs.
Shae, it sounds like you are simply investing a certain amount of money weekly/monthly/quarterly so it would probably be DCA. They rebalancing between stocks/bonds has nothing to do with DCA/VA, that's just how it handles your asset allocation :)
@NLL, would you hold forth on why your DCA v. VA scenario comparison departs so acutely from that of Michael Edleson’s book (chapter 8)? Looks like he used Gaussian distribution for Ch. 8, but he explains that the earlier chapters established VA > DCA using historical market data. You used 40 year timeframe, while Edleson appears to have used 5, 10, and 20 year timeframes.
From my understanding that has to do with the very large bull run we have generally been in for the last number of years causing VA to have large cash piles and missing big returns for 10+ years. And the comparisons of data can draw very very different conclusions based on start and end dates
A comment about the data used. when testing ways to invest one should only use raw data! Don't use data that has had dividend, and/or inflation rates incorporated in them. There is a theory that the current price has all known information discounted into the price already. The same should be true for the historical raw data, if you use data with extra dividend/inflation data added to it you could get an over magnification effect. Also if you decide to use an investment type you will only have the raw data available and your best estimates of growth rate and inflation rates for use.
A comment about the growth rate used, Value Average is highly dependent on using the correct growth rate. Michael Edleson felt that the average growth rate was 12% a year, you used 10% a year this caused "C" to be larger than it needed to be. Also, it would cause Value Averaging to start selling stock especially in the latter part of the testing period. I am not surprised that DCA ended up with the larger account. Also, value averaging is not trying to get you the most in your account, it is trying its hardest to meet your goal of 5,000,000 in 40 years. Also, value averaging is not a set it once and forget it tool. One is supposed to review and adjust the setting on the fly, Michael Edleson recommended once every ten years, myself I would recommend once every five years, using the last ten years of raw data to get a new growth rate and a new value for "C", Michael Edleson explains how to go about doing this in his book in chapter 11.
This video has a major flaw. It assumes that the increase in the portfolio value is the same every month - hence linear. If course your portfolio is going to underperform a DCA when you are aiming for a linear monthly growth. A true comparison and backtesting would require that you set your monthly growth value target to rise exponentially not linearly. Hence the conclusion of this video is wrong.
Nice, lump sum beats DCA which beats VCA, in 70%+ of the time based on total end of day dollar returns - logical ending since stocks go up about 2/3 of the time. May i understand, if the simulation done took into account of: a. interest generated by the unused $ - say if they sat in money markets/CDs or similar, then taken out to buy into VCA b. the risk factor of DCA's return vs VCA's return? something like Risk vs Reward kinda thing like Sharpe ratio? how would it be?
Mun, interest from putting the cash in something like CDs was not considered in this. There are a few reasons for that: First, you generally need to keep the money in CDs for a specific length of time and early withdrawals usually come with some sort of penalty which would wipe out a good chunk of the interest benefit if you had to withdraw early to meet the contribution target. Second, in the vast majority of cases the money that was sitting on the sidelines was not sitting there for very long (rarely more than a year or so to be honest, and often only a couple months). So while building some sort of CD ladder could potentially get around that early withdrawal issue, the benefits for most people in most cases would still likely not be big enough to change the overall conclusions. And third, the current interest rate environment makes it unlikely that an investor taking that approach would be able to bridge the gap between the strategies enough to change the overall conclusions found in the video. So if this had been used it would change some of the specific figures, but the overall conclusions would probably remain largely the same. DCA would outperform VA more often than not and Lump Sum would slightly outperform DCA more often than not. And both VA and DCA gives some additional downside protection relative to the lump sum approach in those rare cases where you're investing at a market peak :) A Sharpe ratio or similar risk-adjusted metric were not calculated as I didn't feel it added much to the overall conversation in this particular video. All three scenarios were investing into the same asset so the risk-adjusted returns would be a lot more heavily dependent on the investing start date than anything else (i.e. VA may have the higher risk-adjusted returns over certain time horizons when investing at or shortly before a market peak, but would likely trail the other approaches in other scenarios). I felt this idea was already mostly covered by looking at the minimum ending net worth figures compared to the average and max figures :) Thanks for the questions!
Gee why don’t you speak faster lol Your drawing is much better than your yak yak yak yaking!!! I love the way you guys always want a like prior to anything!!! Not In this case for sure!!!
From my experience DCA will always win. The best perk of VCA is having extra cash on hand to take advantage of bear markets and underpriced stocks. Good video
I was thinking a modified approach could be the best way to go. Investing all excess money on hand with a lump sum approach but then also DCA on a monthly/biweekly/weekly approach as you earn money. If you are going to be putting money in during a down month why not really take advantage of those lower months when you can!
This is the best way to do it
I do both dca and vca.. 2020 was exceptional year.. Took some off top early in 2021.. Great topic! Thx
I was thinking a modified approach like you use could be the best way to go. If you are going to be putting money in during a down month why not really take advantage of those lower months when you can!
What about the “YOLO everything on meme stocks” approach?
Potentially very high return ceiling and just as low of a potential return floor ;)
TO THE MOOOOOOOOON
You should have mentioned Michael Edleson’s book, that way your audience can go straight to the font and learn about it if they want to. I do think that one advantage to VA is that one will probably spend less money out of their pocket to reach a certain goal, if that's something you want. Also, in Michael's book, VA does beat DCA if you're using quarterly investments, instead of investing monthly.
Why not mix the two options? Set monthly networth target and a minimum investment amount. Even if you are above your networth target you continue adding the minimum, but in extreme times when you are below your networth target. You add extra to keep up with the target. I think this will be riskier, but I am 23 and I am willing to take a higher risk as I have statistically many years left in my life
The conclusion of this video is wrong because it assumes a fixed growth in value each month. Thus you are comparing a linearly growing portfolio against a an exponentially growing portfolio which the DCA always produces. In that case OF COURSE you are always going to end up with a result where the exponentially growing portfolio is going to outperform the linearly growing portfolio over long enough period. This does not require any backtesting to know that result. Setting a target growth that is linear is actually a really dumb thing to do. Why do you use a fixed monthly growth in your model but not an exponential monthly growth where you aim for higher growth each next month? And then the more interesting part is what exponential power do you use to determine by how much you want to increase your target each month. Well you should use the same exponential power that a DCA approach would yield. Hence the way to PROPERLY compare the strategies is to first compute the DCA computation over the past 30 years. Then from that computation, extract the exponential power and use that to figure out where to set your value growth target for each month. Then do the Value averaging computatoin. You will find remarkably different outcome and the Value approach will always outperform the DCA. The video is totally flawed.
Did you use a spreadsheet like Excel with Oracle-Crystalball to do your simulation?
Excellent video, as always.
Glad you enjoyed it :)
Im investing in a robo advisor, Betterment, that recommends a stock/bond ratio depending on time horizon. I can also auto set deposits weekly/monthly/quarterly into the taxable accounts (and they tell me the estimate of how much ill have when the time horizon comes)
If I turn on their auto setting, they'll rebalance the ratio more bond less stock when i get closer to my time horizon. They also have tax loss harvesting. Would this be DCA or value averaging? It's all ETFs.
Shae, it sounds like you are simply investing a certain amount of money weekly/monthly/quarterly so it would probably be DCA. They rebalancing between stocks/bonds has nothing to do with DCA/VA, that's just how it handles your asset allocation :)
This is DCA
@NLL, would you hold forth on why your DCA v. VA scenario comparison departs so acutely from that of Michael Edleson’s book (chapter 8)? Looks like he used Gaussian distribution for Ch. 8, but he explains that the earlier chapters established VA > DCA using historical market data. You used 40 year timeframe, while Edleson appears to have used 5, 10, and 20 year timeframes.
From my understanding that has to do with the very large bull run we have generally been in for the last number of years causing VA to have large cash piles and missing big returns for 10+ years. And the comparisons of data can draw very very different conclusions based on start and end dates
I’d say it depends how much money you have to invest
That can certainly make quite a difference!
I think the best way is the smart pac. You put more money when the market fails down start to 5%
A comment about the data used. when testing ways to invest one should only use raw data! Don't use data that has had dividend, and/or inflation rates incorporated in them. There is a theory that the current price has all known information discounted into the price already. The same should be true for the historical raw data, if you use data with extra dividend/inflation data added to it you could get an over magnification effect. Also if you decide to use an investment type you will only have the raw data available and your best estimates of growth rate and inflation rates for use.
Lump-sum FTW. Dumping as soon as I can!
A comment about the growth rate used, Value Average is highly dependent on using the correct growth rate. Michael Edleson felt that the average growth rate was 12% a year, you used 10% a year this caused "C" to be larger than it needed to be. Also, it would cause Value Averaging to start selling stock especially in the latter part of the testing period. I am not surprised that DCA ended up with the larger account. Also, value averaging is not trying to get you the most in your account, it is trying its hardest to meet your goal of 5,000,000 in 40 years. Also, value averaging is not a set it once and forget it tool. One is supposed to review and adjust the setting on the fly, Michael Edleson recommended once every ten years, myself I would recommend once every five years, using the last ten years of raw data to get a new growth rate and a new value for "C", Michael Edleson explains how to go about doing this in his book in chapter 11.
Great points!
Good job
Thanks 👍
Thanks
You're welcome!
My feeble attempt at engagement
ty
You bet :)
Many think DCA is no good. The truth is: DCAing stupidly is no good! 🙂👍
This video has a major flaw. It assumes that the increase in the portfolio value is the same every month - hence linear. If course your portfolio is going to underperform a DCA when you are aiming for a linear monthly growth. A true comparison and backtesting would require that you set your monthly growth value target to rise exponentially not linearly. Hence the conclusion of this video is wrong.
Nice, lump sum beats DCA which beats VCA, in 70%+ of the time based on total end of day dollar returns - logical ending since stocks go up about 2/3 of the time.
May i understand, if the simulation done took into account of:
a. interest generated by the unused $ - say if they sat in money markets/CDs or similar, then taken out to buy into VCA
b. the risk factor of DCA's return vs VCA's return? something like Risk vs Reward kinda thing like Sharpe ratio?
how would it be?
Mun, interest from putting the cash in something like CDs was not considered in this. There are a few reasons for that:
First, you generally need to keep the money in CDs for a specific length of time and early withdrawals usually come with some sort of penalty which would wipe out a good chunk of the interest benefit if you had to withdraw early to meet the contribution target.
Second, in the vast majority of cases the money that was sitting on the sidelines was not sitting there for very long (rarely more than a year or so to be honest, and often only a couple months). So while building some sort of CD ladder could potentially get around that early withdrawal issue, the benefits for most people in most cases would still likely not be big enough to change the overall conclusions.
And third, the current interest rate environment makes it unlikely that an investor taking that approach would be able to bridge the gap between the strategies enough to change the overall conclusions found in the video.
So if this had been used it would change some of the specific figures, but the overall conclusions would probably remain largely the same. DCA would outperform VA more often than not and Lump Sum would slightly outperform DCA more often than not. And both VA and DCA gives some additional downside protection relative to the lump sum approach in those rare cases where you're investing at a market peak :)
A Sharpe ratio or similar risk-adjusted metric were not calculated as I didn't feel it added much to the overall conversation in this particular video. All three scenarios were investing into the same asset so the risk-adjusted returns would be a lot more heavily dependent on the investing start date than anything else (i.e. VA may have the higher risk-adjusted returns over certain time horizons when investing at or shortly before a market peak, but would likely trail the other approaches in other scenarios). I felt this idea was already mostly covered by looking at the minimum ending net worth figures compared to the average and max figures :)
Thanks for the questions!
Bro, you talk too fast. Slow down
Change playback speed to 0.75, you have the power
We're long overdue for a crypto video my friend.
Maybe Daniel isn't into Crypto
@@shaereub4450He doesn't make videos for himself
Gee why don’t you speak faster lol
Your drawing is much better than your yak yak yak yaking!!! I love the way you guys always want a like prior to anything!!!
Not In this case for sure!!!
🌞☕️☕️
DAMN you talk to fast😂
Neither is the best.