4 ms·

Until the beginning of the 20th century, stocks were viewed as a purely speculative investment. The idea that buy and hold will provide great returns is a moder

Until the beginning of the 20th century, stocks were viewed as a purely speculative investment. The idea that buy and hold will provide great returns is a modern one and is supported by the growth of the stock market in the 20th century. There is also the issue of survivorship bias. The SP500 and Dow Jones indices regularly discard the losers and add new companies, so we don't know the true results of holding companies for a long period of time.

martincmartin 4y ago
> There is also the issue of survivorship bias. The SP500 and Dow Jones indices regularly discard the losers and add new companies, so we don't know the true results of holding companies for a long period of time. That's certainly true in marketing material of fund managers. But it's not true in honest academic papers, like this one. There's lots of data out that about total returns of the stock market, that takes into account dividends and survivorship bias.
bumby 4y ago
Do you happen to know where to find historical data on such holdings? For example, how do I find the historical ETF holdings at a particular point in time?
- Kon-Peki 4y ago
  ETFs? Welllllll, how about mutual funds instead? A lot of ETFs are just exchange-traded mutual funds. Mutual funds are required to periodically report their holdings: https://www.sec.gov/edgar/sec-api-documentation https://www.sec.gov/edgar/sec-api-documentation Rate limit yourself to under 10 requests per second, and put contact info into your user-agent if you'd like them to contact you about problems. > The APIs are updated in real-time as filings are disseminated. The submissions API is updated with a typical processing delay of less than a second; the xbrl APIs are updated with a typical processing delay of under a minute. However these processing delays may be longer during peak filing times.
  - bumby 4y ago
    Thanks, this is useful. However, I was specifically looking for information on the SPDR ETFs.
- f5ve 4y ago
  These data exist but in many cases it is expensive to obtain. Even complete historical data for the SP500 is not a trivial thing to find. (ETF current holdings, I believe, are legally required to be published)
bagacrap 4y ago
Is there a strategy that reliably beats buy and hold? Obviously not or everyone would do it. And no investment professional worth their salt would advise to buy individual companies to hold for the long term (almost all will go out of business or underperform eventually) so I'm not sure what purpose the survivorship bias comment serves.
- jatins 4y ago
  Momentum strategies have done better over long term[0]. Maybe someone else can contribute literature that is more trustworthy than BI Also investing in Index is Momentum strategy disguised as buy and hold since Index will rebalance periodically to add growing companies and remove the lagging ones. [0] https://www.businessinsider.com/momentum-trading-has-beaten-the-pants-off-of-buy-and-hold-2010-9 https://www.businessinsider.com/momentum-trading-has-beaten-...
- kqr 4y ago
  Constant fraction rebalanced beats buy and hold in the long run[1], under a wide variety of market conditions, including adversarial ones. The difficulty lies in finding (a) what fractions are appropriate for the assets you want to invest in, and (b) the appropriate rate of rebalancing, accounting for transaction costs. [1]: I don't have citations at hand, but the papers are reproduced (among many other interesting ones!) in Theory and Practise of the Kelly Capital Growth Investment Criterion.
- trompetenaccoun 4y ago
  Everyone doesn't know how to successfully invest or there wouldn't be so many middle class and poor people. Buying and holding (a broad index tracker) seems like the best strategy for someone who doesn't know about the businesses or can't be bothered to follow the market. It's the most passive strategy. Imo the mistake most make is they mentally compare it to themselves reading a bit online and then picking stocks based on what they feel will perform well. Of course this can only go wrong. If you're an expert in an area like the first quant traders or someone who understands startups and does angel investing, you can absolutely beat the market and people do it all the time. That's how some people get rich, you can not become truly rich by working a 9-5 office job and buying a few stocks. Unless theoretically you trade the riskiest penny stocks with leverage and magically come out on top every time like some of those WSB guys. I've yet to hear of cases where this works long term. The smartest ones know when they get lucky and stop playing, because that's pure gambling.
baby 4y ago
Tangential but why do people always cite the DJ when it only has a dozen handpicked companies? Doesn’t sound like a great metric

martincmartin 4y ago

> There is also the issue of survivorship bias. The SP500 and Dow Jones indices regularly discard the losers and add new companies, so we don't know the true results of holding companies for a long period of time. That's certainly true in marketing material of fund managers. But it's not true in honest academic papers, like this one. There's lots of data out that about total returns of the stock market, that takes into account dividends and survivorship bias.

bumby 4y ago

Do you happen to know where to find historical data on such holdings? For example, how do I find the historical ETF holdings at a particular point in time?

Kon-Peki 4y ago
ETFs? Welllllll, how about mutual funds instead? A lot of ETFs are just exchange-traded mutual funds. Mutual funds are required to periodically report their holdings: https://www.sec.gov/edgar/sec-api-documentation https://www.sec.gov/edgar/sec-api-documentation Rate limit yourself to under 10 requests per second, and put contact info into your user-agent if you'd like them to contact you about problems. > The APIs are updated in real-time as filings are disseminated. The submissions API is updated with a typical processing delay of less than a second; the xbrl APIs are updated with a typical processing delay of under a minute. However these processing delays may be longer during peak filing times.
- bumby 4y ago
  Thanks, this is useful. However, I was specifically looking for information on the SPDR ETFs.
f5ve 4y ago
These data exist but in many cases it is expensive to obtain. Even complete historical data for the SP500 is not a trivial thing to find. (ETF current holdings, I believe, are legally required to be published)

bagacrap 4y ago

Is there a strategy that reliably beats buy and hold? Obviously not or everyone would do it. And no investment professional worth their salt would advise to buy individual companies to hold for the long term (almost all will go out of business or underperform eventually) so I'm not sure what purpose the survivorship bias comment serves.

jatins 4y ago
Momentum strategies have done better over long term[0]. Maybe someone else can contribute literature that is more trustworthy than BI Also investing in Index is Momentum strategy disguised as buy and hold since Index will rebalance periodically to add growing companies and remove the lagging ones. [0] https://www.businessinsider.com/momentum-trading-has-beaten-the-pants-off-of-buy-and-hold-2010-9 https://www.businessinsider.com/momentum-trading-has-beaten-...
kqr 4y ago
Constant fraction rebalanced beats buy and hold in the long run[1], under a wide variety of market conditions, including adversarial ones. The difficulty lies in finding (a) what fractions are appropriate for the assets you want to invest in, and (b) the appropriate rate of rebalancing, accounting for transaction costs. [1]: I don't have citations at hand, but the papers are reproduced (among many other interesting ones!) in Theory and Practise of the Kelly Capital Growth Investment Criterion.
trompetenaccoun 4y ago
Everyone doesn't know how to successfully invest or there wouldn't be so many middle class and poor people. Buying and holding (a broad index tracker) seems like the best strategy for someone who doesn't know about the businesses or can't be bothered to follow the market. It's the most passive strategy. Imo the mistake most make is they mentally compare it to themselves reading a bit online and then picking stocks based on what they feel will perform well. Of course this can only go wrong. If you're an expert in an area like the first quant traders or someone who understands startups and does angel investing, you can absolutely beat the market and people do it all the time. That's how some people get rich, you can not become truly rich by working a 9-5 office job and buying a few stocks. Unless theoretically you trade the riskiest penny stocks with leverage and magically come out on top every time like some of those WSB guys. I've yet to hear of cases where this works long term. The smartest ones know when they get lucky and stop playing, because that's pure gambling.

baby 4y ago

Tangential but why do people always cite the DJ when it only has a dozen handpicked companies? Doesn’t sound like a great metric