1Yr
If you successfully study a STEM subject, you should earn your entire current capital in one month. So apart from the money market reserve, it shouldn't really matter in absolute terms where you invest and how much. So I would suggest that you just try different strategies. ETFs, certificates, B&H, market timing, etc. You can then use the experience gained when money really comes in. Good luck anyway!
•
55
•
@Epi Yes, that's right :). My goal is to study computer science, if possible at Stanford (Computer Sciences program). But that would only be possible with a scholarship. That's why it's still so hard for me to estimate how much I'll have to rely on my savings in a few months.
•
11
•1Yr
@LukasRichter That sounds like a plan! If it works out, you'll need quite a bit of money for your move, vacation, girlfriend (especially in the U.S.), etc. I'd rather play it safe and collect the prime rates in a relaxed manner at $XEOD.
••
Hi @Epi, I wanted to get back to you with a question. I've read about your GTAA and dual momentum strategy in several posts and in your profile and I've looked at the papers on it. I would now like to program a backtesting tool to compare retrospectively how different assets and allocations of them would have performed in the past. Of course, the performance in euros is more interesting for me (at least at the moment :)). However, many ETFs in particular also have a large proportion invested in securities listed in USD. I'm not sure whether I should look at the performance in USD or in EUR for the comparison, as the comparison in EUR naturally involves some currency fluctuations and is therefore more of a form of noise in the data, which is largely decoupled from the respective performance of the asset classes. On the other hand, only the performance in "my" currency counts for me anyway. Do you have any ideas on how I should approach this? Comparison in EUR or USD? Or are there such strong correlations between the development of USD-EUR exchange rates and the asset classes that the exchange rates do not significantly distort the general development and momentum of the assets?
••
1Yr
@LukasRichter Hello Lukas, the currency issue is indeed quite complicated and I have not yet clarified it conclusively for myself. There are many things to consider: 1. if it is only a matter of developing a good model, Portfoliovisualizer is sufficient. There is no corresponding backtesting option for euro securities, if only because the corresponding assets are often not listed for that long. In USD you get good data up to around 2000, in EUR up to 2013. 2. The differences between models in USD and EUR correlate with the currency fluctuations and these can sometimes be 20%pa. There are assets that tend in one direction with a certain USD trend, e.g. gold tends to rise when the USD falls. In these cases, you can simply use hedged ETFs. With other assets, it is the other way around, you can take unhedged ones. You then have to understand the economic logic of the assets in the model. 3. the signal situation is difficult. Sometimes a buy signal is triggered in USD but not in EUR. Here you have to decide which rule to follow. One possibility is to take the signals in USD but trade the ETFs in EUR. But this also has consequences. 4 On the whole, however, it can be said that the performances roughly converge. The currency areas and interest rate policies are very dependent on each other in the long term, so that short-term deviations only play a subordinate role. So you can take a fairly relaxed approach to the matter. I hope this has helped a little. If you have developed a backtest tool, you can get in touch. There is a Signal group that also works on such things.
••
@Epi OK, then I'll probably just use a toggle for signals in EUR or native currency for the backtesting tool. I didn't know about the Portfoliovisualizer tool (the whole site in general) and had pretty much the same thing in mind. It's definitely very exciting, especially with the presets for the various strategies. I'll write again when I'm a bit further along.
•
11
•