Author Topic: Downloadable historical list of various index composition?  (Read 7587 times)

Rmt

  • 5 O'Clock Shadow
  • *
  • Posts: 24
Downloadable historical list of various index composition?
« on: April 30, 2015, 05:38:02 PM »
Hello,

Does anyone know where I can get/download free historical composition of what makes up various stock market indices or ETFs at various points in time?
I tried googling but articles I've found were old with nonworking download urls.
Or if anyone already have this data and can share with me?
I want to backtest some strategies free of survivorship bias...

Thanks!

Rmt

  • 5 O'Clock Shadow
  • *
  • Posts: 24
Re: Downloadable historical list of various index composition?
« Reply #1 on: May 01, 2015, 11:43:17 AM »
Okay, I was able to get what I needed from someone... thanks.

If anyone has a good conservative, active stock investment strategy (besides just investing in VTI ) and would like for me to try out in backtesting, please let me know. You can also use PM features to send me privately. I'm just looking for something to beat VTI by a few percent, NO crazy risky strategies!

Thanks!

forummm

  • Walrus Stache
  • *******
  • Posts: 7389
  • Senior Mustachian
Re: Downloadable historical list of various index composition?
« Reply #2 on: May 01, 2015, 08:06:28 PM »
Okay, I was able to get what I needed from someone... thanks.

If anyone has a good conservative, active stock investment strategy (besides just investing in VTI ) and would like for me to try out in backtesting, please let me know. You can also use PM features to send me privately. I'm just looking for something to beat VTI by a few percent, NO crazy risky strategies!

Thanks!

Is the historical list something you can share with others? If you're trying to beat VTI by a few percent, you will be more likely in the long run to lose to VTI. Just be prepared for that.

AZryan

  • Stubble
  • **
  • Posts: 152
Re: Downloadable historical list of various index composition?
« Reply #3 on: May 01, 2015, 11:54:17 PM »
"-I'm just looking for something to beat VTI by a few percent, NO crazy risky strategies!"

Replace VO for VTI. It's the Mid-Cap index. A bit more extremes, but follows the total market 95% and makes ~1.4% more than VTI in the long run (and actually handled the horrible 2000 tech crash better). Small Cap index's about the same, but I prefer the Mid 'cuz they're all pretty damn big billion dollars companies/known brands (Seagate, Chipolte, Harley, etc.).

Forummm's wrong about this ...
"-If you're trying to beat VTI by a few percent, you will be more likely in the long run to lose to VTI."

VO doesn't lose to VTI in the long run. I think Forummm's probably thinking 'Active stock picking "gurus" vs. Blind Indexing', but VO is indexing same as VTI is.

This link might help, too...

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults

From as far back as it goes ('72 on), you can see Mid Caps beat Total Market.

forummm

  • Walrus Stache
  • *******
  • Posts: 7389
  • Senior Mustachian
Re: Downloadable historical list of various index composition?
« Reply #4 on: May 02, 2015, 09:18:43 AM »
"-I'm just looking for something to beat VTI by a few percent, NO crazy risky strategies!"

Replace VO for VTI. It's the Mid-Cap index. A bit more extremes, but follows the total market 95% and makes ~1.4% more than VTI in the long run (and actually handled the horrible 2000 tech crash better). Small Cap index's about the same, but I prefer the Mid 'cuz they're all pretty damn big billion dollars companies/known brands (Seagate, Chipolte, Harley, etc.).

Forummm's wrong about this ...
"-If you're trying to beat VTI by a few percent, you will be more likely in the long run to lose to VTI."

VO doesn't lose to VTI in the long run. I think Forummm's probably thinking 'Active stock picking "gurus" vs. Blind Indexing', but VO is indexing same as VTI is.

This link might help, too...

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults

From as far back as it goes ('72 on), you can see Mid Caps beat Total Market.

I assumed from OP's statements that since he was looking for individual stocks in the indexes, he was looking to do something like active stock picking. You were less assumptive in your response--good for you. You are correct that during certain time periods, certain asset classes have outperformed large caps. It's unclear whether this trend will continue. Bogle thinks it won't (http://www.vanguard.com/bogle_site/sp20020626.html), but he could easily be wrong. You are trading some stability for some hope of increased return. If you are in the asset accumulation phase, that can help increase your long-term portfolio yield (if there is outperformance of VTI). If you are in retirement, the added volatility could deplete your assets faster. It's a risk/reward issue. I think holding VO is a reasonable strategy. Maybe it works. Or maybe it doesn't and you get VTI performance (or a little worse). You won't go bust.

AZryan

  • Stubble
  • **
  • Posts: 152
Re: Downloadable historical list of various index composition?
« Reply #5 on: May 02, 2015, 12:59:19 PM »
Quote from: forummm
I assumed from OP's statements that- -he was looking to do something like active stock picking.

Yeah, I didn't catch that in his 2nd post. Only bothered with the original question. My bad.

Quote from: forummm
-during certain time periods, certain asset classes have outperformed large caps. It's unclear whether this trend will continue. Bogle thinks it won't-

You linked to a Bogle speech from 2002 noting Small Caps vs. Large Cap, rather than Mids. But he does note (now in 2015) Small beat Large for over 80 years. You would have made more money if you put all your Large Cap into Small Cap right after he made that speech saying that they really don't do better -not that that means anything. Just funny, I think.

You can see for yourself that the Mid Index follows the Total market by ~95%. This is true daily, weekly, monthly, 3, 5, 10 years. The link I posted shows this going back 42 years.

Quote from: forummm
You are trading some stability for some hope of increased return.

As I noted... yes. But 'slight' volatility, for a MUCH increased return that's been historically proven long term. It's as much of a sure thing as the idea that the stock market always goes up. Sometimes bonds do better. But mostly... no, stocks win. And Mids (Small) beat Large.

Quote from: forummm
If you are in the asset accumulation phase, that can help increase your long-term portfolio yield (if there is outperformance of VTI).


It will increase it. An historical, long term fact.

Quote from: forummm
If you are in retirement, the added volatility could deplete your assets faster.

That's true of anything that 'could be'. This becomes the same argument as someone choosing 100% stocks vs. having a significant % of bonds in the mix. The latter will, long term, likely reduce your total. The former takes being adaptive and on your toes much more if things go south early on.

If you had 100% VO or VTI in retirement... going by a blind 4% rule in 1972, you'd survive with either one. But 1M in '72 would be 5.6M+ now with VTI. But would be 40M+ with VO! That's HUGE for just slightly more volatility. Pick most any random smaller segment of time and you get similar results. Just have to be on your toes in the early days (like we all should be anyway).

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1397
  • Age: 40
  • Location: Texas
Re: Downloadable historical list of various index composition?
« Reply #6 on: May 02, 2015, 05:26:06 PM »
Recreate this strategy here but do it for S&P 500 stocks, Russell 2000/3000 stocks, or other indexes of your choice.

http://bettersystemtrader.com/bollinger-band-trading-strategy/


Sent from my iPhone using Tapatalk

forummm

  • Walrus Stache
  • *******
  • Posts: 7389
  • Senior Mustachian
Re: Downloadable historical list of various index composition?
« Reply #7 on: May 02, 2015, 05:43:47 PM »
Quote from: forummm
I assumed from OP's statements that- -he was looking to do something like active stock picking.

Yeah, I didn't catch that in his 2nd post. Only bothered with the original question. My bad.

Quote from: forummm
-during certain time periods, certain asset classes have outperformed large caps. It's unclear whether this trend will continue. Bogle thinks it won't-

You linked to a Bogle speech from 2002 noting Small Caps vs. Large Cap, rather than Mids. But he does note (now in 2015) Small beat Large for over 80 years. You would have made more money if you put all your Large Cap into Small Cap right after he made that speech saying that they really don't do better -not that that means anything. Just funny, I think.

You can see for yourself that the Mid Index follows the Total market by ~95%. This is true daily, weekly, monthly, 3, 5, 10 years. The link I posted shows this going back 42 years.

Quote from: forummm
You are trading some stability for some hope of increased return.

As I noted... yes. But 'slight' volatility, for a MUCH increased return that's been historically proven long term. It's as much of a sure thing as the idea that the stock market always goes up. Sometimes bonds do better. But mostly... no, stocks win. And Mids (Small) beat Large.

Quote from: forummm
If you are in the asset accumulation phase, that can help increase your long-term portfolio yield (if there is outperformance of VTI).


It will increase it. An historical, long term fact.

Quote from: forummm
If you are in retirement, the added volatility could deplete your assets faster.

That's true of anything that 'could be'. This becomes the same argument as someone choosing 100% stocks vs. having a significant % of bonds in the mix. The latter will, long term, likely reduce your total. The former takes being adaptive and on your toes much more if things go south early on.

If you had 100% VO or VTI in retirement... going by a blind 4% rule in 1972, you'd survive with either one. But 1M in '72 would be 5.6M+ now with VTI. But would be 40M+ with VO! That's HUGE for just slightly more volatility. Pick most any random smaller segment of time and you get similar results. Just have to be on your toes in the early days (like we all should be anyway).

Some good points. As I said, it's a decent strategy. Apparently mid-caps did underperform large caps for the 90's and then outperformed for the decade after that. So there will be some up times and down times relative to large caps.

http://www.forbes.com/sites/rickferri/2013/04/29/no-free-lunch-from-equal-weight-sp-500/

forummm

  • Walrus Stache
  • *******
  • Posts: 7389
  • Senior Mustachian
Re: Downloadable historical list of various index composition?
« Reply #8 on: May 02, 2015, 05:50:33 PM »
Recreate this strategy here but do it for S&P 500 stocks, Russell 2000/3000 stocks, or other indexes of your choice.

http://bettersystemtrader.com/bollinger-band-trading-strategy/


Sent from my iPhone using Tapatalk

Would you say this has similar risks (and rewards) as the dual momentum strategy on the other thread? Whipsaw, flash crashes, etc?

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1397
  • Age: 40
  • Location: Texas
Re: Downloadable historical list of various index composition?
« Reply #9 on: May 02, 2015, 06:13:58 PM »
It introduces individual stock risk, but it's a trend following system so it acts similar. Most importantly though it's uncorrelated to a diversified trend following system so using them in conjunction with each other works better than each of the individual systems. I have a diversified system I like, but still working on a long-only individual stock trend following system to combine with it.


Sent from my iPhone using Tapatalk

AZryan

  • Stubble
  • **
  • Posts: 152
Re: Downloadable historical list of various index composition?
« Reply #10 on: May 03, 2015, 09:58:17 AM »
Quote from: forummm
Apparently mid-caps did underperform large caps for the 90's and then outperformed for the decade after that.

You mean they basically matched or OUTperformed for 8 out 10 years in the 90's, and then OUTperformed for most of the next 15 years(-to-date), AND the multiple decades before the 90's, as well.

But feel free to cherry pick a TWO-year period where they underperformed. They also fell harder in '08 if you want to cherry pick some more.
I'd rather accept those rare losses, and collect more money in the long run. Obviously, for something that's more volatile. it has to be down below Large Caps/Total Market sometimes. But this is far easier to accept than the understanding we all have that the market itself will have periods of being down.

In any reasonable sense of the long term investing, Mid Caps are better AND with lower adjusted risk.

I don't know what I was supposed to get from your Rick Ferri link? -same as your last link to Bogle's hugely long speech that never mentioned Mid Caps. In the Bogle link, he absolutely failed to understand the illogic of Pascal's Wager -which he tried to use to illustrate his own well-reasoned logic in examining the market, so it actually lowered my regard for his advice.

Maybe you could just pick out the key point and post it, rather than make someone guess where inside a long article we should look to be shown how we're wrong.

Here's a good article on Mid Caps. The whole article directly relates to my point, but the conclusion says it all -"-Mid-cap stocks as an asset class have represented a time-tested investing “sweet spot” in the stock market, consistently outperforming both large- and small-cap stocks on both an absolute and risk-adjusted basis.-"

https://www.americancentury.com/content/dam/americancentury/direct/rd/pdf/Mid_Cap_Stocks_The_Markets_Sweet_Spot_-_Direct.pdf

forummm

  • Walrus Stache
  • *******
  • Posts: 7389
  • Senior Mustachian
Re: Downloadable historical list of various index composition?
« Reply #11 on: May 03, 2015, 11:55:06 AM »
Quote from: forummm
Apparently mid-caps did underperform large caps for the 90's and then outperformed for the decade after that.

You mean they basically matched or OUTperformed for 8 out 10 years in the 90's, and then OUTperformed for most of the next 15 years(-to-date), AND the multiple decades before the 90's, as well.

But feel free to cherry pick a TWO-year period where they underperformed. They also fell harder in '08 if you want to cherry pick some more.

Dude, take it easy. No one is arguing with you or criticizing you. I said several times that you had good points and that what you suggested was a reasonable strategy. I also provided more information for the OP to learn from--since he started the thread and wants to learn more about investing. Clearly you are very well informed and don't need any information from me. But there are others reading that may not be as expert as you. If they are taking your advice, they should know that occasionally it will underperform (which was the point of the link I provided and should have been clear from my post). Sometimes people don't expect a strategy to underperform, and then panic and sell low. People should go into investing knowing more or less what to expect. If the decade of performance that Ferri mentions can be explained by 2 years, that's interesting information (and he should have included it in his article). If it was just 2 years, then there must have been quite a disparity for those 2 years. Again, someone investing in this way should be aware of those occasional events and not freak out when they happen.

We're all friends here.

skyrefuge

  • Handlebar Stache
  • *****
  • Posts: 1007
  • Location: Suburban Chicago, IL
Re: Downloadable historical list of various index composition?
« Reply #12 on: May 04, 2015, 01:04:26 PM »
Okay, I was able to get what I needed from someone... thanks.

heh...what's with the vague secrecy? Is there some sort of pirated database that people are passing around secretly via private messages or something? Is there an investor-torrentz.com website I'm unaware of?!

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1397
  • Age: 40
  • Location: Texas
Re: Downloadable historical list of various index composition?
« Reply #13 on: May 04, 2015, 01:17:23 PM »
Usually you gotta pay money for a historical list of index constituents. And if you want historical price data on those companies, especially the delisted ones, that can be pretty expensive. Fundamental data at the historical point in time on top of that? Even more expensive.

Here's a list of S&P 500 joiners and leavers from 1994 to 2013. To even get this list I had to get it from someone with a paid subscription to an expensive service.

https://drive.google.com/file/d/0BzyyTlvGE-T2S09pSS14ZjU0V3M/view?usp=sharing
« Last Edit: May 04, 2015, 01:20:56 PM by hodedofome »

skyrefuge

  • Handlebar Stache
  • *****
  • Posts: 1007
  • Location: Suburban Chicago, IL
Re: Downloadable historical list of various index composition?
« Reply #14 on: May 04, 2015, 01:41:53 PM »
Usually you gotta pay money for a historical list of index constituents. And if you want historical price data on those companies, especially the delisted ones, that can be pretty expensive. Fundamental data at the historical point in time on top of that? Even more expensive.

I guess that makes sense. It's just an unusual case, particularly in today's Internet, since it's all data that is (was) publicly available; the providers are getting lots of money for information that anyone could have collected. But, "anyone" didn't, so good for them for charging whatever the market will bear for the effort they put into saving that data. Sounds like a good business to be in! (until someone starts up a web-based open-source competitor!) I wonder if anything beyond a license agreement is violated by sharing such public data? Anyway, thanks for sharing that bit that you did!

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1397
  • Age: 40
  • Location: Texas
Re: Downloadable historical list of various index composition?
« Reply #15 on: May 04, 2015, 01:58:03 PM »
I guess that makes sense. It's just an unusual case, particularly in today's Internet, since it's all data that is (was) publicly available; the providers are getting lots of money for information that anyone could have collected. But, "anyone" didn't, so good for them for charging whatever the market will bear for the effort they put into saving that data. Sounds like a good business to be in! (until someone starts up a web-based open-source competitor!) I wonder if anything beyond a license agreement is violated by sharing such public data? Anyway, thanks for sharing that bit that you did!

Quandl is doing the best it can to provide as much free data as possible. The data they charge for is because it cost them money to get it. $100/month for an individual https://www.quandl.com/

AZryan

  • Stubble
  • **
  • Posts: 152
Re: Downloadable historical list of various index composition?
« Reply #16 on: May 04, 2015, 03:38:11 PM »
Quote from: forummm
-If it was just 2 years, then there must have been quite a disparity for those 2 years. Again, someone investing in this way should be aware of those occasional events and not freak out when they happen.

Your advice seems very sensible, but it's actually backwards and misguided. There was nothing to freak out about. There was a very rare divergence between the Mids and Large Caps in the last 2 years of the 90's, but both were still going up during that time. Newbies weren't freaking out and selling either index.

This weird discrepancy in the long(almost any)-term tight correlation was because the Large Caps bubbled, then popped in the worst crash any one of us probably ever saw.

So, it was the Large Caps that went bonkers, not the Mids. The 2000's are called the 'lost decade' in investing, but they weren't 'lost' for the Mids -nor were the 90's, 80's, 70's, etc.

I'm claiming now that a Mid Index is straight-up safer/less risky, makes more $, AND doesn't require any special advice other than exactly the same thing that you'd tell anyone in the Total Market Index.

I think that's the one point we don't agree on. And I could be wrong. I want to be corrected if I am. No one else here advocates for Mids as their core (or only) holding like they do with the Total Market (as far as I've ever seen).

Lack of diversification seems like the strongest argument against a Mid Index (usually ~400 or so companies), but most of us here (including me) consider the Total Market and the 500 Index to be 99% interchangeable, so having WAY less than the thousands of companies in the Total Market obviously isn't an inherent problem.

Rmt

  • 5 O'Clock Shadow
  • *
  • Posts: 24
Re: Downloadable historical list of various index composition?
« Reply #17 on: May 05, 2015, 12:26:26 AM »
Is the historical list something you can share with others? If you're trying to beat VTI by a few percent, you will be more likely in the long run to lose to VTI. Just be prepared for that.

The data I got turned out to be dirty. but I did find iShares site, you can download the ETF composition going back to 9 years... so I am using that as historical index composition.
If you know of any other sites that allows one to download their historical ETF composition, please let me know!

devan 11

  • 5 O'Clock Shadow
  • *
  • Posts: 54
  • Location: Iowa, USA
Re: Downloadable historical list of various index composition?
« Reply #18 on: May 05, 2015, 07:46:37 AM »
 Kenneth Fisher's "The Wall Street Waltz" always teaches me new lessons of investing each time I read it.   The visual charts give a long view of history of investing.   Don't buy the ebook version though.  Too much detail is lost from the coarse pictures of charts.  As an aside, Ken Fisher gets high marks by independent review of whether his recommendations make money.  I bogle, but if I had to go to fee advise, he is among the three best in my list.