Applying trinity study/firecalc/cfiresim in own country

[k290]

firecalc and such simulators are based on US historical  data. This is not useful to people in my country, since my country has a different economic climate, different political influences, different economic principles, different taxes, different changes in inflation rates, different imports/exports etc. This means I need to apply firecalc related studies to my own country's data.

I have spent hours and hours looking for historical data, reading equations, figuring out how the trinity study is applied etc. I am now frustrated due to:

• Lack of readily available historical data
• Difficulty understanding relevant US vernacular e.g. Treasury Bonds and converting it to an equivalent (or near equivalent) term in my country, in order to research it
• Difficulty understanding the Trinity Studies motivations behind the data chosen e.g. S&P 500, after the war etc.
• Difficulty figuring out how to apply historical data that I have found, correctly

In summary I am not a finance major or researcher. I do not work in finance. My understanding of these things is so limited to begin with. My fantastic idea of building a firecalc-like website for people in my own country to use have dwindled away.

(I'm sure, given two years and the amount of time available I had when I was studying my post-grad, university resources and a prof. to guide, I could probably do it. Unfortunately I can't go back to school and do a major in finance just so I can do my own Trinity Study.)

Does anyone have any suggestions or alternatives?

UPDATE: I was reminded of a post that arebelspy put on my journal: http://www.advisorperspectives.com/newsletters14/Does_International_Diversification_Improve_Safe_Withdrawal_Rates.php
It seems like whereever Wade Pfau got the data from is the answer. If anyone has insights on how he calculated the SWRs/where the data is from please post  and it will save me a lot of reading:)

[hodedofome]

Don't have access to the data you require but I will say this - one of the main ways to determine the robustness of a strategy is if it can survive being used in other asset classes. Meaning, if a strategy works well in one country, but doesn't work in other countries, it should make you question how robust the strategy really is. It is quite possible the 4% rule is curve fit to only US data (I haven't read the complete study and I haven't seen all the data myself, just thinking out loud).

[forummm]

If you just own the whole world stock index (or multiple indexes that together make up the whole world) you don't have the risk of your home country's market lagging the rest of the world. You do have some currency related risk. But those currencies will be pretty stable ones (USD, EUR, CAD, etc), so that should limit some of the consequences.

[deborah]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

[forummm]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

Well the US is 55% of the whole world right now. And the other developed countries are something like 35% more I think. So it's quite heavy in the highly correlated developed world. US and EAFE performed about the same over the last 40 years.

The currency risk, especially for an unknown currency, is a harder question. I think the currency risk for very stable currencies is minimal over the long run. If you own EUR denominated securities through a USD denominated index, you don't really see much change from currency fluctuations, even with the recent 20% drop  (or whatever it was) in the EUR. To be a little simplistic, that's because the stock prices rose in EUR at the same time the EUR fell, because the companies earn a lot in USD and also become proportionally cheaper in USD so USD funds move over to buy them, which increases them to about the same level relative to USD.

[deborah]

The AUD has dropped from $1.10US (or thereabouts) to $0.70US (or thereabouts) in the past couple of years. That is much more significant than you were thinking of. Admittedly that gives and Australian currency advantage for prior investments. However, the worst years differ for different countries and all it takes is for the World Index to go down in sync with the home currency going up (which happened here in recent deborah memory - see graph attached) for a much lower return.

You can see from the graph (for example) that for Australians, the 2008 crash meant we could not get the deals you would have been getting on whole world shares because our $ was much lower than the US$ at that time.

[StockBeard]

Does anyone have any suggestions or alternatives?

I used to live in Japan. Before that I was in France, now I'm in the US.
My solution has been to follow the same investment rules as someone in the US, it's just a matter of investing in the same funds. In lieu of bonds, I have a US Bond ETF from Schwab, for example. In most countries, you should be able to invest in these US ETFs, or equivalent.

[k290]

one of the main ways to determine the robustness of a strategy is if it can survive being used in other asset classes. Meaning, if a strategy works well in one country, but doesn't work in other countries, it should make you question how robust the strategy really is.

Yes, the idea is to test whether a strategy works in countries other than the US i.e. my home country. For that you need non-US data.
If firecalc reports 97% success for a particular strategy on US data, it could very well report 94% success if applied to another country.
It is a simulation run over data. By changing the data you are guaranteed with high probability to change the results of the simulation.

I found this in one of the comments on a mmm post: http://www.mrmoneymustache.com/2012/05/29/how-much-do-i-need-for-retirement/

“Wade Pfau (source for much of the data cited above) was the first to put this idea to the test. He applied a research methodology similar to Bengen but with the critical change of using historical international data instead of U.S. data. The results were alarmingly different.

In other words, similar methodology as 2nd Generation models + different data = dramatically different conclusions.

Using 109 years of data for each of 17 different developed countries Pfau determined that a 4% withdrawal rate with a fixed 50/50 asset allocation would have failed in all 17 countries. Yes, a 100% failure rate. You could have run out of money before you ran out of life using the conventional assumptions on foreign country data.

[k290]

Update: I've found the studies by Wade Pfau which includes my home country: South Africa, thanks to a post by arebelspy
In theory I could create a mini-cfiresim for my home country based on wherever he got his data: http://www.advisorperspectives.com/newsletters14/Does_International_Diversification_Improve_Safe_Withdrawal_Rates.php

[lauren_knows]

Update: I've found the studies by Wade Pfau which includes my home country: South Africa, thanks to a post by arebelspy
In theory I could create a mini-cfiresim for my home country based on wherever he got his data: http://www.advisorperspectives.com/newsletters14/Does_International_Diversification_Improve_Safe_Withdrawal_Rates.php

I now see your interest in getting on board with the cFIREsimOpen project. :)

[hodedofome]

From the article:

Clients can expect better outcomes with international diversification, but even with this greater diversification, there is still a chance that 4% will not work and adjustments to spending patterns or asset allocations will be needed in retirement.

This is always been my reason for working part-time in retirement. For 1) people shouldn't sit around doing nothing and 2) we may need the $$$

[deborah]

It would be interesting to see what the outcomes actually were with international diversification. What you would need is the currency movements as well as the world wide data. We tend to assume that international diversification is good, but the figures could actually say the opposite.

[dungoofed]

Subscribing

[Due South]

Hi K290, I have run a rough calc using SA equity returns from 1960and assuming a portfolio of 100% equities. A 3.4% drawdown gives a 98% success rate, and 3% gives 100% success. But I don't place too much weight behind the results using return data going way back  Giventhe political and subsequent economic changes since early '90s I don't think those early returns are representative. Our equity Market is also very narrowly defined and back then was heavily resource/gold focused.  This is changing quite significantly now and going forward.  It's tricky as we have a lot of economic and policyuncertainty going forward and this makes planning (especially over long timeframes) very hard.  Best to diversy portfolio offshore as far as possible. 

[k290]

Hi K290, I have run a rough calc using SA equity returns from 1960and assuming a portfolio of 100% equities. A 3.4% drawdown gives a 98% success rate, and 3% gives 100% success. But I don't place too much weight behind the results using return data going way back  Giventhe political and subsequent economic changes since early '90s I don't think those early returns are representative. Our equity Market is also very narrowly defined and back then was heavily resource/gold focused.  This is changing quite significantly now and going forward.  It's tricky as we have a lot of economic and policyuncertainty going forward and this makes planning (especially over long timeframes) very hard.  Best to diversy portfolio offshore as far as possible.

Interesting. Thanks for creating an account just to post that. That is helpful:)

[Interest Compound]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

The general rule of thumb, is that currency risk over the long-term is 0, but can create swings up or down in the short term. This was analyzed in the following post:

http://forum.mrmoneymustache.com/investor-alley/australiausa-mustachian-philosophy-differences/msg568543/#msg568543

"AUD/USD doubled, dropped about 40%, rose back up, then dropped about 35%.  What impact did that make on the MSCI world index returns had you held in either AUD or USD?  Not much.  This is not evidence that currency risk isn't significant, it's evidence that relatively large currency swings can have a negligible difference."

[johnny847]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

The general rule of thumb, is that currency risk over the long-term is 0, but can create swings up or down in the short term. This was analyzed in the following post:

http://forum.mrmoneymustache.com/investor-alley/australiausa-mustachian-philosophy-differences/msg568543/#msg568543

"AUD/USD doubled, dropped about 40%, rose back up, then dropped about 35%.  What impact did that make on the MSCI world index returns had you held in either AUD or USD?  Not much.  This is not evidence that currency risk isn't significant, it's evidence that relatively large currency swings can have a negligible difference."

I was under the impression that this is not true, at least based on economic principles. There is no economic theory that states that the exchange rate should remain in the long term average zero.

My very simplistic counterexample is the following:
The exchange rate is driven by foreign demand of your country's goods. The more in demand your country's goods are, the more valuable your currency.
If your country's goods are constantly in high demand, then the exchange rate isn't going to average out to zero in the long run.


Again, very simplistic example and I don't have a background in economics. If somebody can prove me wrong I'm all for it.

As for past performance, you can look at past performance all you want, but
1) That case study was only for one pair of currencies.
2) Past performance is not indicative of future returns.


EDIT: Or maybe a better counterexample. Suppose the USD constantly inflates at 2% and the EUR constantly inflating at 3% (pick whatever numbers you want, so long as the EUR constantly inflates faster). I don't see how the exchange rate is going to be constant in the long run.

Furthermore, any government can step in to manipulate their currency however they please.

[Interest Compound]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

The general rule of thumb, is that currency risk over the long-term is 0, but can create swings up or down in the short term. This was analyzed in the following post:

http://forum.mrmoneymustache.com/investor-alley/australiausa-mustachian-philosophy-differences/msg568543/#msg568543

"AUD/USD doubled, dropped about 40%, rose back up, then dropped about 35%.  What impact did that make on the MSCI world index returns had you held in either AUD or USD?  Not much.  This is not evidence that currency risk isn't significant, it's evidence that relatively large currency swings can have a negligible difference."

I was under the impression that this is not true, at least based on economic principles. There is no economic theory that states that the exchange rate should remain in the long term average zero.

My very simplistic counterexample is the following:
The exchange rate is driven by foreign demand of your country's goods. The more in demand your country's goods are, the more valuable your currency.
If your country's goods are constantly in high demand, then the exchange rate isn't going to average out to zero in the long run.


Again, very simplistic example and I don't have a background in economics. If somebody can prove me wrong I'm all for it.

As for past performance, you can look at past performance all you want, but
1) That case study was only for one pair of currencies.
2) Past performance is not indicative of future returns.


EDIT: Or maybe a better counterexample. Suppose the USD constantly inflates at 2% and the EUR constantly inflating at 3% (pick whatever numbers you want, so long as the EUR constantly inflates faster). I don't see how the exchange rate is going to be constant in the long run.

Furthermore, any government can step in to manipulate their currency however they please.

The idea that currency risk goes to 0 over the long-term, isn't based on past performance and looking at charts. Currency risk is bidirectional and expected to be long-term zero. In theory, in the long run, currency values should equalize based on things like the "law of one price".

There are many counter-arguments to this. Paul Marsh, emeritus professor of finance at London Business School, says "Over the very long run, currency barely matters. The trouble with that is that not many investors have 112-year horizons, they need to be extremely patient for that."

Simply looking at Japan's currency vs the US over the last 60 years is a great example of how long you'd have to wait for things to equalize (if they ever do):



Personally, when I see this, it makes me want to diversify more in foreign currencies. Countries tend to fall vs the world faster than they rise, and I'd hate to be on the losing end of that. Especially if I were living somewhere like Australia, with their comparatively tiny and undiversified two-sector economy. It seems much more risky to not invest internationally for someone in that situation.

[johnny847]

However, the whole world has two problems. Firstly, it is lower than the USA figures. Secondly (as you point out) there is currency related risk, which is not always trivial. I have yet to see a study which identifies whether a whole world stock index in general compounds or nullifies the highs and lows of the currency related risk for any country.

The general rule of thumb, is that currency risk over the long-term is 0, but can create swings up or down in the short term. This was analyzed in the following post:

http://forum.mrmoneymustache.com/investor-alley/australiausa-mustachian-philosophy-differences/msg568543/#msg568543

"AUD/USD doubled, dropped about 40%, rose back up, then dropped about 35%.  What impact did that make on the MSCI world index returns had you held in either AUD or USD?  Not much.  This is not evidence that currency risk isn't significant, it's evidence that relatively large currency swings can have a negligible difference."

I was under the impression that this is not true, at least based on economic principles. There is no economic theory that states that the exchange rate should remain in the long term average zero.

My very simplistic counterexample is the following:
The exchange rate is driven by foreign demand of your country's goods. The more in demand your country's goods are, the more valuable your currency.
If your country's goods are constantly in high demand, then the exchange rate isn't going to average out to zero in the long run.


Again, very simplistic example and I don't have a background in economics. If somebody can prove me wrong I'm all for it.

As for past performance, you can look at past performance all you want, but
1) That case study was only for one pair of currencies.
2) Past performance is not indicative of future returns.


EDIT: Or maybe a better counterexample. Suppose the USD constantly inflates at 2% and the EUR constantly inflating at 3% (pick whatever numbers you want, so long as the EUR constantly inflates faster). I don't see how the exchange rate is going to be constant in the long run.

Furthermore, any government can step in to manipulate their currency however they please.

The idea that currency risk goes to 0 over the long-term, isn't based on past performance and looking at charts. Currency risk is bidirectional and expected to be long-term zero. In theory, in the long run, currency values should equalize based on things like the "law of one price".

There are many counter-arguments to this. Paul Marsh, emeritus professor of finance at London Business School, says "Over the very long run, currency barely matters. The trouble with that is that not many investors have 112-year horizons, they need to be extremely patient for that."

Simply looking at Japan's currency vs the US over the last 60 years is a great example of how long you'd have to wait for things to equalize (if they ever do):



Personally, when I see this, it makes me want to diversify more in foreign currencies. Countries tend to fall vs the world faster than they rise, and I'd hate to be on the losing end of that. Especially if I were living somewhere like Australia, with their comparatively tiny and undiversified two-sector economy. It seems much more risky to not invest internationally for someone in that situation.

Oh don't get me wrong, I'm all for investing internationally.

But I've got to agree with Professor Marsh here. Saying currency risk is going to average out to zero in the long run isn't a helpful statement when "long run" can exceed an investor's lifetime. Unless that investor's goal is to leave a large inheritance to their kids.


It is worth noting that historically currency movements have been uncorrelated to movements in stock prices (you can find this in Vanguard's whitepaper on how much international exposure should you hold). But of course you can find exceptions - and of note, Australia is one such exception.
And again, this is past performance.


If you want to remove the currency risk deborah, you can invest in currency hedged international funds. I have not done research on them though so I don't know if there are any good ones with low expenses. Currency hedging does increase costs.

[deborah]

While currency MOVEMENTS are historically 0 (I agree with that), the 4% withdrawal, cFireSim and the trinity study are looking at how things line up in retirement to give you best/worst returns. The graph I gave illustrates the difference currency exchange rates AT THE TIME can make on whole of world share price. If I had bought in the low at 2008, I would have paid DOUBLE my normal rate for a whole of world index because our dollar went down so much with respect to the US dollar for the duration of the problems. It then came back up again, so if I had bought in the dip and sold a year later, when the share prices recovered, I would get (let's say) half the gains that someone in the US doing exactly the same thing would get. This SURELY has a marked effect in the SWR for and Australian buying whole of world shares as against someone from the US.

[Interest Compound]

While currency MOVEMENTS are historically 0 (I agree with that), the 4% withdrawal, cFireSim and the trinity study are looking at how things line up in retirement to give you best/worst returns. The graph I gave illustrates the difference currency exchange rates AT THE TIME can make on whole of world share price. If I had bought in the low at 2008, I would have paid DOUBLE my normal rate for a whole of world index because our dollar went down so much with respect to the US dollar for the duration of the problems. It then came back up again, so if I had bought in the dip and sold a year later, when the share prices recovered, I would get (let's say) half the gains that someone in the US doing exactly the same thing would get. This SURELY has a marked effect in the SWR for and Australian buying whole of world shares as against someone from the US.

Funny how you can see that your market only crashed HALF AS MUCH as the rest of the world in 2008 (a 25% drop instead of a 50% drop), and turn it into a bad thing, in a conversation about the 4% safe withdrawal rate :-P

I understand your main point though. I'd be much more worried about currency risk if my country only represented 2% of the world economy. What if next time currency risk goes the other way, instead of cushioning the crash (turning a 50% world crash into a 25% crash for you), it exacerbates it (by turning a 50% world crash into a 75% crash for you)? Luckily there's a very simple step you can take to mitigate this, use Vanguard's hedged international ETF:

https://www.vanguardinvestments.com.au/institutional/jsp/investments/etf-detail/etfdetailVGADHE.jsp

[deborah]

While currency MOVEMENTS are historically 0 (I agree with that), the 4% withdrawal, cFireSim and the trinity study are looking at how things line up in retirement to give you best/worst returns. The graph I gave illustrates the difference currency exchange rates AT THE TIME can make on whole of world share price. If I had bought in the low at 2008, I would have paid DOUBLE my normal rate for a whole of world index because our dollar went down so much with respect to the US dollar for the duration of the problems. It then came back up again, so if I had bought in the dip and sold a year later, when the share prices recovered, I would get (let's say) half the gains that someone in the US doing exactly the same thing would get. This SURELY has a marked effect in the SWR for and Australian buying whole of world shares as against someone from the US.

Funny how you can see that your market only crashed HALF AS MUCH as the rest of the world in 2008 (a 25% drop instead of a 50% drop), and turn it into a bad thing, in a conversation about the 4% safe withdrawal rate :-P

I understand your main point though. I'd be much more worried about currency risk if my country only represented 2% of the world economy. What if next time currency risk goes the other way, instead of cushioning the crash (turning a 50% world crash into a 25% crash for you), it exacerbates it (by turning a 50% world crash into a 75% crash for you)? Luckily there's a very simple step you can take to mitigate this, use Vanguard's hedged international ETF:

https://www.vanguardinvestments.com.au/institutional/jsp/investments/etf-detail/etfdetailVGADHE.jsp

I don't think you are getting my point at all.

This thread is about applying cFireSim in your own country. I see there being two elements to this.

Firstly, you need the data for your own country, so cFireSim runs against your own country's data as against the US data for the % you have decided to go for homegrown stocks.

Then, you need whole of world data for the % you have decided to go for whole of world stocks (because if, as you say, your country only has 2% of the world stocks, you need some non-homegrown stuff).

Then you need the currency differential to multiply the whole of world stocks so that it gives a true idea of how the whole of world stock prices move for your own country.

Then you need to say what you are doing - let's say 50/50 home/world or whatever, and let the Trinity study or cFireSim or whatever run to see what you actually SHOULD be doing, and whether whole of world REALLY gives an advantage, or whether it is better to stick (even in an environment which is 2% of the world) with just home stocks. And whether the introduction of whole of world stocks actually makes a difference to the 3.6% SWR that is currently quoted for Australia (for example).

The original outcome of the Trinity study is  based on US home stocks. What REALLY happens when you add whole of world into the mix? I don't know of any study that has looked at this. We are ALL being told to diversify, and that diversification is a good thing. If it is, a Trinity study or a cFireSim should show this.

[happy]

Looks like another few years work for Wade and Bo-Knows….

[forummm]

Regarding development, I think it would be not too difficult to adapt cFIREsim to use the data from other countries. The hard part of development is all the other stuff.

However, this means you actually have to have all of the data from other countries, along with an idea of what the interface and option changes will be. And there are a lot of other countries out there. And not all of them have great public data sets of stock market and bond returns and inflation rate going back 150 years.

If someone is really serious about this, I suggest making bo_knows's job a lot easier and locating international data for him. I would suggest that this will be much harder than you think. Perhaps just European data would do the job? Other than a handful of countries (maybe Canada and Australia), that's the only set of nations that was really investible during that time period. And WWI and WWII will be interesting time periods to review.

[deborah]

Well, Canada and Australia have two of the most mustachians per head of population here (after the US) so these countries would probably be very interested in this. (Most bang for Bo's buck)

[dungoofed]

Watching the AUD dive again with this whole Greece thing. They're saying it is not due to AUD not being a "safe haven" currency (which it is), but instead related to expected decreased demand for resources due to a pending slowdown in Europe.

Great. My home currency weakens every time there's a significant sale on shares. Oh well, at least there should be some bargains in the Aussie resource sector.

[lauren_knows]

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

[johnny847]

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

I've always wondered bo_knows, where do you get your data from? It dates all the way back to 1871. The DJIA is the oldest index that I know of, which only started in 1896. And it's a pretty terrible index, only consisting of 30 stocks.

[forummm]

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

I've always wondered bo_knows, where do you get your data from? It dates all the way back to 1871. The DJIA is the oldest index that I know of, which only started in 1896. And it's a pretty terrible index, only consisting of 30 stocks.

http://www.econ.yale.edu/~shiller/data/ie_data.xls

[k290]

Regarding development, I think it would be not too difficult to adapt cFIREsim to use the data from other countries. The hard part of development is all the other stuff.

However, this means you actually have to have all of the data from other countries, along with an idea of what the interface and option changes will be. And there are a lot of other countries out there. And not all of them have great public data sets of stock market and bond returns and inflation rate going back 150 years.

If someone is really serious about this, I suggest making bo_knows's job a lot easier and locating international data for him. I would suggest that this will be much harder than you think. Perhaps just European data would do the job? Other than a handful of countries (maybe Canada and Australia), that's the only set of nations that was really investible during that time period. And WWI and WWII will be interesting time periods to review.

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

Well, yes, this was the reason I started this thread to begin with: where to get the data??!! :)  [And consistent data at that]

I eventually found access to consistent data for over 15 countries. Unfortunately i have to buy a license for it. For now we may only be using data for South Africa (my home country) as a proof of concept [and because it is useful to me personally]. But need to talk to a sales rep to see if I can get only a subset of the data instead of all of it.

I haven't been through all the code, but from the UI, I believe the social security part of cfiresim is too US specific and may need to be turfed for an international fork of the project. It is also possible that deep within the code there is a formula or two that may only really correctly apply to the US. e.g. I haven't had a look at what cfiresim actually does with the pension amounts provided. Depending on what it does, these equations could vary from country to country. If tax is taken into account anywhere in the equations, that could automatically be moot for several countries with different tax law. Just my thoughts of what could make it require more than just throwing new data at it, (since I haven't had a look at what's actually going on in there.)

I'm a bit busy with studying at the moment but I'll update here if I can get a subset of the data as getting it for all 15+ countries costs a lot of money.

[Xlar]

Regarding development, I think it would be not too difficult to adapt cFIREsim to use the data from other countries. The hard part of development is all the other stuff.

However, this means you actually have to have all of the data from other countries, along with an idea of what the interface and option changes will be. And there are a lot of other countries out there. And not all of them have great public data sets of stock market and bond returns and inflation rate going back 150 years.

If someone is really serious about this, I suggest making bo_knows's job a lot easier and locating international data for him. I would suggest that this will be much harder than you think. Perhaps just European data would do the job? Other than a handful of countries (maybe Canada and Australia), that's the only set of nations that was really investible during that time period. And WWI and WWII will be interesting time periods to review.

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

Well, yes, this was the reason I started this thread to begin with: where to get the data??!! :)  [And consistent data at that]

I eventually found access to consistent data for over 15 countries. Unfortunately i have to buy a license for it. For now we may only be using data for South Africa (my home country) as a proof of concept [and because it is useful to me personally]. But need to talk to a sales rep to see if I can get only a subset of the data instead of all of it.

I haven't been through all the code, but from the UI, I believe the social security part of cfiresim is too US specific and may need to be turfed for an international fork of the project. It is also possible that deep within the code there is a formula or two that may only really correctly apply to the US. e.g. I haven't had a look at what cfiresim actually does with the pension amounts provided. Depending on what it does, these equations could vary from country to country. If tax is taken into account anywhere in the equations, that could automatically be moot for several countries with different tax law. Just my thoughts of what could make it require more than just throwing new data at it, (since I haven't had a look at what's actually going on in there.)

I'm a bit busy with studying at the moment but I'll update here if I can get a subset of the data as getting it for all 15+ countries costs a lot of money.

Do you have a link to where we could buy this data? We could maybe looking into buying access by pooling our funds for the betterment of the community. Of course, the main hurdle would probably be the licensing terms and conditions.

[johnny847]

Regarding development, I think it would be not too difficult to adapt cFIREsim to use the data from other countries. The hard part of development is all the other stuff.

However, this means you actually have to have all of the data from other countries, along with an idea of what the interface and option changes will be. And there are a lot of other countries out there. And not all of them have great public data sets of stock market and bond returns and inflation rate going back 150 years.

If someone is really serious about this, I suggest making bo_knows's job a lot easier and locating international data for him. I would suggest that this will be much harder than you think. Perhaps just European data would do the job? Other than a handful of countries (maybe Canada and Australia), that's the only set of nations that was really investible during that time period. And WWI and WWII will be interesting time periods to review.

The data is the primary concern.  If you gave me year on year data, I could easily integrate it.  I've just had problems finding reliable data before the 1970s, for anything outside of the US.

Well, yes, this was the reason I started this thread to begin with: where to get the data??!! :)  [And consistent data at that]

I eventually found access to consistent data for over 15 countries. Unfortunately i have to buy a license for it. For now we may only be using data for South Africa (my home country) as a proof of concept [and because it is useful to me personally]. But need to talk to a sales rep to see if I can get only a subset of the data instead of all of it.

I haven't been through all the code, but from the UI, I believe the social security part of cfiresim is too US specific and may need to be turfed for an international fork of the project. It is also possible that deep within the code there is a formula or two that may only really correctly apply to the US. e.g. I haven't had a look at what cfiresim actually does with the pension amounts provided. Depending on what it does, these equations could vary from country to country. If tax is taken into account anywhere in the equations, that could automatically be moot for several countries with different tax law. Just my thoughts of what could make it require more than just throwing new data at it, (since I haven't had a look at what's actually going on in there.)

I'm a bit busy with studying at the moment but I'll update here if I can get a subset of the data as getting it for all 15+ countries costs a lot of money.

Do you have a link to where we could buy this data? We could maybe looking into buying access by pooling our funds for the betterment of the community. Of course, the main hurdle would probably be the licensing terms and conditions.

Right. It's fairly likely that if you have to buy the data, you can't just use it in a free to use website like cFIREsim.

[LazySod]

Sorry for bumping this old thread, i just think it's still relevant for many people outside the US.
I am from Brazil and i was researching this subject when I found this thread.
Our stock market is kinda new but there is 45 years of data available for free on the web. Same for inflation and fixed income investments.
Not sure if that is enough for such study.
If you guys ever consider adding Brazil to the tool I would be glad to help.

[Tyler]

Nice dig.  This topic is a passion of mine.

Since this thread was first started I developed my own tool for calculating withdrawal rates using data from other countries.  So far it can handle the US, Canada, Australia, Germany, and the UK, but I plan to add more countries in the future.  It also can calculate returns for broad international indices and convert the results to local currencies and inflation.  Check it out:

https://portfoliocharts.com/portfolio/withdrawal-rates/

BTW, all of the comments here about calculation and data challenges are on-point.  It's a complicated topic! 

[DavidAnnArbor]

I noticed that the international stock markets are up this year quite a bit more than the US markets, and from what I read it's largely because the dollar has gone down in value compared to other currencies.