As technology becomes more prominent in the industry, and as financial instruments become increasingly more complex - fundamental analysis, and thus the CFA, will fall into obscurity. Lucky are those of us with a solid mathematics/science/programming background!
well played sir
Pixel?
at the end of the day, fundamental analysis is what is required. programmers simply build spreadsheets, but what to do with the information obtained is more important. you can always hire a programmer to make models and spreadsheets. as an example, the financial crisis we saw brewed on the backs of complex financial instruments. however, a person that had no understanding of it and smart enough to stay out of things they did not understand did not lose millions. i take a history background over a computer programming one anyday. programmers are basiclaly tools used. nice try nerd.
Actually, it’s probably best to have a solid fundamental background in addition to a solid fundamental background.
FrankArabia Wrote: ------------------------------------------------------- > at the end of the day, fundamental analysis is > what is required. > > programmers simply build spreadsheets, but what to > do with the information obtained is more > important. > > you can always hire a programmer to make models > and spreadsheets. > > as an example, the financial crisis we saw brewed > on the backs of complex financial instruments. > however, a person that had no understanding of it > and smart enough to stay out of things they did > not understand did not lose millions. > > i take a history background over a computer > programming one anyday. programmers are basiclaly > tools used. > > nice try nerd. nice frank
Yes, computers are perfectly able to predict human emotions. I guess that explains why they predicted the housing bubble so well. Naturally, all of the QA models knew that the stock market was going to fall over 50%. That’s why there wasn’t a single QA fund that lost money. In fact, all of them rose 100% by shorting the market in a 100% perfect position. Furthermore, it’s why computers have replaced psychiatrists and drugs. Ever hear of the acronym GIGO? For a supposed quant jock you’re unsurprisingly dense.
Mathematics degrees from the University of Phoenix don’t count dude!
Supreme, just sign up for L1 again in June.
spierce Wrote: ------------------------------------------------------- > Yes, computers are perfectly able to predict human > emotions. Can humans? What’s your argument? > I guess that explains why they predicted the > housing bubble so well. Naturally, all of the QA > models knew that the stock market was going to > fall over 50%. That’s why there wasn’t a single > QA fund that lost money. In fact, all of them > rose 100% by shorting the market in a 100% perfect > position. 100% short was far from perfect, and fyi if the market falls 50%, you don’t make 100%. > Furthermore, it’s why computers have replaced > psychiatrists and drugs. Please keep your personal issues out of this. > Ever hear of the acronym GIGO? The underwear??
FrankArabia Wrote: ------------------------------------------------------- > at the end of the day, fundamental analysis is > what is required. Both quantitative and fundamental analysis are required. You might have been able to get a job 15 years ago without quantitative skills, but it’s nearly impossible today. > programmers simply build spreadsheets, but what to > do with the information obtained is more > important. > > you can always hire a programmer to make models > and spreadsheets. > > as an example, the financial crisis we saw brewed > on the backs of complex financial instruments. > however, a person that had no understanding of it > and smart enough to stay out of things they did > not understand did not lose millions. Wow, just wow. Do you really think it was the person who understood these models that was leveraging this garbage 30 to 1? > i take a history background over a computer > programming one anyday. programmers are basiclaly > tools used. > > nice try nerd. I too value history highly, but you’re way off the mark here.
justin88 Wrote: ------------------------------------------------------- > spierce Wrote: > -------------------------------------------------- > ----- > > Yes, computers are perfectly able to predict > human > > emotions. > > Can humans? What’s your argument? But computers are programed by humans. > > > > I guess that explains why they predicted the > > housing bubble so well. Naturally, all of the > QA > > models knew that the stock market was going to > > fall over 50%. That’s why there wasn’t a > single > > QA fund that lost money. In fact, all of them > > rose 100% by shorting the market in a 100% > perfect > > position. > > 100% short was far from perfect, and fyi if the > market falls 50%, you don’t make 100%. I realize that. > > > > Furthermore, it’s why computers have replaced > > psychiatrists and drugs. > > Please keep your personal issues out of this. > They aren’t my personal issues. It’s your issue you can’t see the fallacies of your arguments. > > > Ever hear of the acronym GIGO? > > The underwear?? Heh. You’re supposedly an advocate of a quant yet you don’t know the most basic premise of modeling? When it comes down to it TA is nothing more than looking in the rearview mirror for where you’re going to go, not where you’ve been. Considering that you cannot predict the unpredictable by utilizing models, a model is only as good as it’s universe of variables. Since you cannot program in human emotions or any “Black Swan” events, your model can’t predict much at all. Sure, FA can’t do that either, but at least FA would let you estimate how a company would perform under those events, how the market performs is immaterial since the premise of FA investment isn’t market performance. If anything, the only reason why TA works is because others think it works, creating a market of its own, taking funds from other TA players. Sure, high-speed trading is TA but it is fundamentally different than what is practiced by all but the biggest banks. It is a stacked deck since only but those players have access to those systems or pools of liquidity for placing trades ahead of the market. Finally, they utilize sub-pennying, again, an unfair advantage. As far as your comment above about models being misunderstood by those using them. I would contend that the models were flawed in the first place. Especially considering there wasn’t a single model that took into account broad-based price declines in housing. GIGO.
Quant models really can’t handle regime changes very well. Sure, you can program them after the fact to adjust for a regime change, but you can’t look forward. Humans are able to forward-look for changes better than machines can, at least at present, and very likely for a long time to come. That doesn’t mean that humans can do it perfectly, but it was clear to many that the housing market was set for a bubble, but there was really no way for a machine to pick up on that. Quantitative investing will have a part to play, and it will tend to excel when the world economy is relatively calm and unturbulent and politics and regulation is less likely to change the rules of the economy and investment. However, when volatility is very low, it is sensible to back off of quant models, because it is very likely that many quant people are in the same trade, which means that your positions are riskier than they look from past data. If your positions are riskier than they look by traditional, it means you’re overexposed unless you’ve explicitly compensated for the hidden risks.
Fundamental people are those who cause a crisis, quants fail to detect them and make them worse.
justin88 Wrote: ------------------------------------------------------- > FrankArabia Wrote: > -------------------------------------------------- > ----- > > at the end of the day, fundamental analysis is > > what is required. > > Both quantitative and fundamental analysis are > required. You might have been able to get a job > 15 years ago without quantitative skills, but it’s > nearly impossible today. > > > > programmers simply build spreadsheets, but what > to > > do with the information obtained is more > > important. > > > > you can always hire a programmer to make models > > and spreadsheets. > > > > as an example, the financial crisis we saw > brewed > > on the backs of complex financial instruments. > > however, a person that had no understanding of > it > > and smart enough to stay out of things they did > > not understand did not lose millions. > > Wow, just wow. Do you really think it was the > person who understood these models that was > leveraging this garbage 30 to 1? > > > > i take a history background over a computer > > programming one anyday. programmers are > basiclaly > > tools used. > > > > nice try nerd. > > I too value history highly, but you’re way off the > mark here. So i’m valuing China Mobile here. What “quant” skills do I need. I got the annual reports, conference calls, company website info and information on the chinese communications industry. In addition, i’m looking at their competitors as well. Do i even need to use a spreadsheet other than to keep the data nice and clean and perhaps change the discount or growth rate? please explain how quant skills will help me make my decision to buy more or sell this security. I admit however, that I rather not build the model and instead would like someone to put all this stuff in a spreadsheet for me so i can just sit and think. some fancy VBA would be nice but highly unlikely it will contribute to my decision. thanks
mcpass Wrote: ------------------------------------------------------- > Fundamental people are those who cause a crisis, > quants fail to detect them and make them worse. What’s so fundamental about asset bubble?
Lol, I’m well aware of garbage in, garbage out. Everyone remotely intelligent knows this. Of course fundamental analysis suffers from the same problem; annual reports, government numbers, conference calls, etc. each have their own significant biases. Gleaning actual information from anything is difficult for anyone, fundies and quants alike. It seems there is little understanding of quantitative analysis on this board, which is not surprising given its secretive nature. First, QA rarely tries to value single stocks independently, instead looking at relative value (compare with technical analysis). Quantitative analysis actually shines in turbulent times; even though the models are imperfect, they usually fair better than their fundamental counterparts, for a variety of reasons (eg. they tend to have lower beta risk and are less sensitive to individual event risk). As an added plus, quant strategies (particularly the ones collecting liquidity premiums) tend to perform well with high volumes found during panics and manias. Perhaps most importantly, quantitative analysis allows one to trade a lot more consistently. One can test idea(s) more rigorously and can scale idea(s) in many dimensions (# names, # of trades, trade execution, etc), and thus take a larger amount of smaller bets with the same amount of capital. Any trader can explain why this is desirable.
^Aren’t most quant strategies a formalized sets of trading rules taken from TA and FA? i.e. their value is in being able to process and transact on much more data quickly while sticking to these rules formally and without emotion, whereas the same can’t be expected of a trader
I’m not sure about “most” per se, but those are a subset, yes. Part of the value comes from fast execution and non-discretionary decision making, but unless you’re among the GETCOs of the world, there is almost always a large alpha component as well. adavydov7 Wrote: ------------------------------------------------------- > ^Aren’t most quant strategies a formalized sets of > trading rules taken from TA and FA? i.e. their > value is in being able to process and transact on > much more data quickly while sticking to these > rules formally and without emotion, whereas the > same can’t be expected of a trader
adavydov7 Wrote: ------------------------------------------------------- > ^Aren’t most quant strategies a formalized sets of > trading rules taken from TA and FA? i.e. their > value is in being able to process and transact on > much more data quickly while sticking to these > rules formally and without emotion, whereas the > same can’t be expected of a trader Yes. What is a “quantitative” model anyway? If you run a regression to determine some fundamental analysis variable, you just used a quantitative model…