6k views

I think you need to differentiate between Q-quants vs P-quants. The former might not use Econometrics, but P-quants use them a lot.

2k views

Heston gives an expression for the characteristic function, from which option prices can be computed. Therefore it can be calibrated (statically) on a set of vanilla option prices with different ...

2k views

In 1983 he was using Hidden Markov Models. Now he employs 100+ PhDs, therefore I expect he will have 50+ strategies using 200+ predictors. And set up as a production line, from the teams importing and ...

2k views

FO is shrinking across the large investment banks. The market is not developing new products that will need new pricing formulas, if anything it is reverting to more vanilla structures. Nowdays FO ...

2k views

A factor model has the form $$r_{j,t}=\sum_n \beta_{j,n} f_{n,t}+\epsilon_{j,t}$$ Where $r_{j,t}$ is the return of stock $j$ at time $t$, $\beta_{j,n}$ is the sensitivity (factor loading) of stock $j$ ...

624 views

I don't see this issue in Bloomberg, therefore I would assume it is just bad data in your source. Based on my snaps I suspect that your bid-ask is 15 min delayed. This stock is very active today, ...

vote
252 views

VG belongs in the family of variance-mean mixture models. Given a horizon $T$ the distribution of log-returns $f$ is a mixture of Gaussians $f_G$ with randomised mean and variance. The randomisation ...

442 views

I would say that $\log K/F$ points towards a log-normal type model. If I were you I would experiment with the moneyness defined as $K-F$ instead. This would make it consistent with normal dynamics. ...

201 views

The first process is a BM. The second does not exist in continuous time. The variance goes down too slowly with dt and the process blows up at the limit. You can break the (0,1) interval into 1, 100,...

667 views

You can guesstimate by vega weighted implied vol. This is why: Say that you have a portfolio of options with prices $P_j$. Each one of them has a different pricing function $f_j$ (as function of vol) ...

745 views

I personally use the simple Garch(1,1) for volatility filtering in the risk management area. In fact in most cases I don't even estimate the parameters, I stick 0.94 for mean reversion, 0.04 for the ...

684 views

It is not the fact that volatility is time varying that creates the skew per se, but the fact that volatility is negatively correlated with the spot. That is to say, as the stock/index price declines ...

9k views

I think to gain intution you have to understand that the same agents that value the stocks will value the options. And agents compensate for volatility by demanding higher expected returns. Therefore ...

394 views

Regression analysis, as a minimization of the sum of squared errors, does not require normality of the error term. The requirements are that errors are homoscedastic and uncorrelated. And these are ...

2k views
If by 'solve' you mean how do we know that $\ln S_t$ is the right change of variable, then you can go by the following (not rigorous) line of thought: Ito's fomula suggests that given an SDE $$dX_t = ... View answer 2 answers votes 890 views Accepted answer 6 votes I believe that the confusion arises because of the wrong treatment of NIG. The answer to the question you link is misleading, as it simulates under P which is not appropriate for option pricing. None ... View answer 1 answers vote 117 views 5 votes What happened was the BoJ announcement. Such large scale news are well covered in mainstream media (ft, bloomberg, etc) and also mainstream anti-media (eg zerohedge). View answer 1 answers votes 223 views Accepted answer 5 votes My understanding, in that context, is that signal indicates that you want to hold a share (signal is 1) or hold no shares (signal is zero). Therefore taking the diff will tell you if you want to buy (... View answer 3 answers votes 44k views 5 votes Assuming that you are working for a bank, there are three different P&amp;Ls depending on the function/ usage: Actual P&amp;L calculated by Finance/ Product Control and is based on the actual price ... View answer 1 answers votes 3k views Accepted answer 5 votes Intuition: You can think of the vol smile as a reflection of the risk neutral distribution (compared to the Black Scholes Gaussian density). A fat tailed distribution creates the smile: fat tail -> ... View answer 4 answers votes 10k views 5 votes The concept of 'mean reversion' is tricky in continuous time. Most people would call 'mean reverting' a process where the drift pulls back towards a long run mean, and I assume that this is what you ... View answer 1 answers votes 134 views 4 votes The price of a derivative does not explicitly depend on the expected return of the underlying, however the price change or return of the derivative depends on the return of the underlying. Hence the ... View answer 3 answers votes 229 views 4 votes IMO: Volatility is a risk factor not an asset class. Asset classes are collections of assets and volatility is not one. Options, volatility derivatives, etc, are asset classes which might offer ... View answer 5 answers votes 651 views 4 votes I'd use FFT or similar rather than direct integration. Here is an old paper with Heston example: Option pricing using fractional FFT View answer 1 answers vote 1k views 4 votes The ADF test assumes the DGP$$ \Delta y_t = \alpha +\beta t +\gamma y_t +\delta_1 \Delta y_{t-1}+\cdots +\delta_k \Delta y_{t-k}+\epsilon_t $$The parameters are estimated using OLS on a sample of ... View answer 3 answers votes 4k views 4 votes Large? ? The relationship between normal and log returns is$$(normal return) = exp(log return)-1$$Therefore log-returns can be from -\infty to +\... View answer 6 answers votes 3k views 4 votes "But just for fun, let's say Pr(S1=Su)=1% and Pr(S1=Sd)=99%, in which case, on average, the call at time 1 would be worth 0.01*10 = 0.1. How would anyone be willing to pay 9.28 for that ? I'm ... View answer 3 answers votes 2k views 4 votes B1~N(0,1) and B2=B1+Z, for Z~N(0,1). From that E(B1*B1)=E(B1*B2)=1, E(B2*B2)=2. Therefore they are bivariate Gaussian with covariance matrix (1,1;1,2) therefore probability is around 12%, which is the ... View answer 2 answers votes 320 views 4 votes There is a shortcut around the Forward Equation when you are looking for the stationary distribution. Let me write$$ dX = \mu(X)dt +\sigma(X)dW $$for$$ \mu(x)=b(1-x)-ax\ \text{ and }\ \sigma^2(x)...