Seriously, is VIX broken or some crazy stuff?!

jys78 said:
Do you guys even know what the VIX actually is? What it represents and how it is -calculated-?
More...

Exactly. +1

If you're going to trade VIX derivatives or products based on them, make sure you know exactly what VIX actually is and how it's calculated: https://www.cboe.com/tradable_products/vix/

Just because you predict SPX will go down over a given time period, that doesn't mean you should just knee-jerk and buy a VIX derivative. You can get horribly burned even if you are right about your prediction on SPX. If you predict SPX will go down then use SPX derivatives instead, plain and simple. VIX is not inverse SPX!
 
Statistical Trader said:
Exactly. +1

If you're going to trade VIX derivatives or products based on them, make sure you know exactly what VIX actually is and how it's calculated: https://www.cboe.com/tradable_products/vix/

Just because you predict SPX will go down over a given time period, that doesn't mean you should just knee-jerk and buy a VIX derivative. You can get horribly burned even if you are right about your prediction on SPX. If you predict SPX will go down then use SPX derivatives instead, plain and simple. VIX is not inverse SPX!
More...
Correct, the correlation is about negative 75%
 
stochastix said:
Correct, the correlation is about negative 75%
More...

Absolutely correct.

-82% daily over last 100 days
-78% daily over last 250 days
-73% daily over last 1000 days

Of course other time frames (eg: weekly) and lookback lengths or rolling window lengths will give different results. And of course this is just standard, out-of-the-box correlation, nothing fancy.

I've read some of what @stochastix has written on ET in the past and he is definitely way beyond the standard correlation stuff... but it seems like some people don't even know these basics. Learn it or lose money, what else can I say?

---
RealTest Code:

Import:
DataSource: Yahoo
IncludeList: SPY
IncludeList: ^VIX
StartDate: Earliest
EndDate: Latest
SaveAs: SPY_VIX_Yahoo.rtd

Data:
VIXClose: Extern($^VIX, Close)
VIXChg: PctChg(VIXClose, 1)
SPYClose: Extern($SPY, Close)
SPYChg: PctChg(SPYClose, 1)
R_SPX_VIX_100: Correl(SPYChg, VIXChg, 100)
R_SPX_VIX_250: Correl(SPYChg, VIXChg, 250)
R_SPX_VIX_1000: Correl(SPYChg, VIXChg, 1000)

Scan:
ClosePrice: Close
VIXClose: VIXClose
SPYClose: SPYClose
R_SPX_VIX_100: R_SPX_VIX_100
R_SPX_VIX_250: R_SPX_VIX_250
R_SPX_VIX_1000: R_SPX_VIX_1000

EDIT: I used SPY as an approximation to SPX due to a data import error I didn't want to spend time fixing.
 
Last edited:
This is elite trader. Plug some head and shoulders and wedges and you can trade anything.



Statistical Trader said:
Exactly. +1

If you're going to trade VIX derivatives or products based on them, make sure you know exactly what VIX actually is and how it's calculated: https://www.cboe.com/tradable_products/vix/

Just because you predict SPX will go down over a given time period, that doesn't mean you should just knee-jerk and buy a VIX derivative. You can get horribly burned even if you are right about your prediction on SPX. If you predict SPX will go down then use SPX derivatives instead, plain and simple. VIX is not inverse SPX!
More...
 
SoyUnGanador said:
Market is rocketing to the downside like I called it, but VIX is actually DOWN today? Its at 27 despite the market being down so bigly over last few months and days. Feels like it should be at like 40 by now at least.

What gives? Is that an indication the market thinks there is not much downside left? Or if there something else more complicated going on!?!?!
More...

RE UVXY etf : -8% ytd seems to be pretty good for an etf that has gone from $1.76 Billion to $11 since 2011 :).
 
Last edited:
Statistical Trader said:
Absolutely correct.

-82% daily over last 100 days
-78% daily over last 250 days
-73% daily over last 1000 days

Of course other time frames (eg: weekly) and lookback lengths or rolling window lengths will give different results. And of course this is just standard, out-of-the-box correlation, nothing fancy.

I've read some of what @stochastix has written on ET in the past and he is definitely way beyond the standard correlation stuff... but it seems like some people don't even know these basics. Learn it or lose money, what else can I say?

---
RealTest Code:

Import:
DataSource: Yahoo
IncludeList: SPY
IncludeList: ^VIX
StartDate: Earliest
EndDate: Latest
SaveAs: SPY_VIX_Yahoo.rtd

Data:
VIXClose: Extern($^VIX, Close)
VIXChg: PctChg(VIXClose, 1)
SPYClose: Extern($SPY, Close)
SPYChg: PctChg(SPYClose, 1)
R_SPX_VIX_100: Correl(SPYChg, VIXChg, 100)
R_SPX_VIX_250: Correl(SPYChg, VIXChg, 250)
R_SPX_VIX_1000: Correl(SPYChg, VIXChg, 1000)

Scan:
ClosePrice: Close
VIXClose: VIXClose
SPYClose: SPYClose
R_SPX_VIX_100: R_SPX_VIX_100
R_SPX_VIX_250: R_SPX_VIX_250
R_SPX_VIX_1000: R_SPX_VIX_1000

EDIT: I used SPY as an approximation to SPX due to a data import error I didn't want to spend time fixing.
More...


Wow, I never realized you were a Wizard, Statistical Trader. That is awesome.
 
You must log in or register to reply here.
Back
Top