Six Figure Investing - Latest Comments

Re: Dividend Capture Strategies

John Hanekamp — Wed, 19 Jun 2019 18:47:21 -0000

SDS is 2x leverage

Re: Trading in IRA accounts, and avoiding “free riding”

Vance Harwood — Sun, 16 Jun 2019 09:50:25 -0000

None that I'm aware of.

Re: Trading in IRA accounts, and avoiding “free riding”

Nagaraj Chekka — Fri, 14 Jun 2019 16:50:01 -0000

Are there any other brokers other than TDAmeritrade and IB, where this T+2 rule is waived ?

Re: Private: Goodbye VXX, Hello VXXB

Joel — Thu, 23 May 2019 22:48:12 -0000

Hey thanks, for some reason it showed up with some blank rows yesterday so I thought it wasn't properly recombined. Looks ok now.

Re: Predicting Stock Market Returns—Lose the Normal and Switch to Laplace

Vance Harwood — Thu, 23 May 2019 15:10:38 -0000

No, the standard normal distribution has a mean of 0 and a standard deviation of 1 the standard lognormal distribution has a mean of 1 and a standard deviation of 1 but in practice, the mean can be anywhere depending on the process. In general, any time you have a repetitive, relatively stable multiplicative process you'll end up with a log-normal distribution just like an additive process will end up with a normal distribution.

Re: Private: Goodbye VXX, Hello VXXB

Vance Harwood — Thu, 23 May 2019 13:50:24 -0000

Hi Joel, The Yahoo finance data looks valid to me from VXX inception in 2009 until now. Since VXXB was identical to VXX they just used that data for the time that VXX wasn't trading. If you want data before 2009 I have a simulation that goes back to March of 2004. https://sixfigureinvesting.... -- Vance

Re: Private: Goodbye VXX, Hello VXXB

Joel — Thu, 23 May 2019 08:32:00 -0000

Hi Vance, where can we get past prices of VXX now? Seems like Yahoo already overwritten the data for VXX ticker. Thanks.

Re: Predicting Stock Market Returns—Lose the Normal and Switch to Laplace

Ashler — Tue, 21 May 2019 14:58:21 -0000

Wouldn’t a lognormal distribution imply strictly positive returns?

Re: Going short on VIX?

Vance Harwood — Tue, 21 May 2019 12:03:57 -0000

Hi Mark, ZIV's asset level is at ~ $80 million which is a relatively healthy level, probably profitable for the issuer, Credit Suisse. The drop in assets is probably due to the recent market downturn, shorter term traders tend to exit short volatility funds during downturns--when volatility is usually elevated and the VIX futures are not in strong contango. -- Vance

Re: Going short on VIX?

Mark — Tue, 21 May 2019 11:51:58 -0000

Vance,
I was referring to the number of "shares" outstanding. Yesterday there were 1,088,000, two weeks ago there was 1,413,000.

Thanks for your response, Mark

Re: Going short on VIX?

Vance Harwood — Tue, 21 May 2019 11:17:37 -0000

Hi Mark, I don't understand your question. ZIV is only down to 70 from a recent high of 75. Can you clarify? -- Vance

Re: Going short on VIX?

Mark — Tue, 21 May 2019 05:00:00 -0000

Vance,

Do you have any ideas why the outstanding ZIV ETNs has been dropping so much the last few weeks. Is it in danger of going away?

Thanks

Re: Private: Precisely Forecasting Stock Price Ranges with Volatility

Wilson — Wed, 08 May 2019 21:06:53 -0000

More than happy to share my thought process. Money management is key to maintaining a positive skew for a system idea i've got, and being able to more closely account for the fat tails in intraday ranges is something the Laplace seems suited for. So for every trade generating a median forecast then +/- 1,2,3 standard deviation levels to take profit and run trailing stops on from the initiation of a trade would be the aim.

So i was thinking of using Average True Range as an estimator of vol, then feeding it into the equivalent of:

Pn = Ps * e(GMcc*n+k*stdev*square root (n))
Pn = Ps* e(GMcc*n)
Pn = Ps * e(GMcc*n-k*stdev*square root (n))
for the Laplace distribution

What's unclear to me is how i adjust this to the appropriate form for laplace.
On critical levels i was just wondering if the empirical rule/ 66-95-99.7 confidence levels for a laplace distribution exist, in a normal distribution it's just 2,4 and 6 sigma from the mean.

would i need to use the inverse function and feed in the probabilities ? not sure how i would adjust sigma in that case.
I kind of understand what you mean about calibrating the volatility term. That'd be multiplying it by some amount to ensure that say 5 sigma occurs with the same frequency as in the data.
Thanks heaps

Re: Private: Precisely Forecasting Stock Price Ranges with Volatility

Vance Harwood — Wed, 08 May 2019 13:21:39 -0000

I do think it makes sense to use the laplace for the intraday moves. You'll have to scale the volatility metric correctly to account for your shorter time period. I don't know what you mean by "critical values", can you say more? -- Vance

Re: Predicting Stock Market Returns—Lose the Normal and Switch to Laplace

Vance Harwood — Wed, 08 May 2019 13:15:30 -0000

HI J.
I believe the difference between the two approaches is that I was looking at daily returns, and they are looking at monthly returns. Since monthly returns are a multiplicative process (series of daily returns multiplied together) I would expect the monthly returns to have a lognormal distribution not a laplacian one (result of central limit theorem). The JP Morgan paper is interesting because it shows correlation effects--the tendency for market declines to be faster than you would predict from a random walk process. I think the key benefit of both of these analysis is that it shows that using a standard normal distribution of returns is way too optimistic when evaluating risk.

Re: Private: Precisely Forecasting Stock Price Ranges with Volatility

Wilson — Wed, 08 May 2019 05:03:53 -0000

Hi Vance,
Thanks for putting up such an interesting post. I'm trying to build a take profit/stop loss model for the intraday timescale, just wondering if this makes sense. If i was to use the laplace distribution to generate the x standard deviation envelopes you're using here, how stable/known are the critical values? If you've seen anything like this i would be even more grateful if you could point me in the right direction.
Many Thanks
will

Re: Predicting Stock Market Returns—Lose the Normal and Switch to Laplace

Paul — Fri, 03 May 2019 18:29:13 -0000

Hello Vance, Thank you for sharing your research in this blog. Your ideal scale factor of 1.19 is for daily returns. How would the ideal scale factor be modified for say, weekly or monthly returns?

Re: Predicting Stock Market Returns—Lose the Normal and Switch to Laplace

J. Natael — Tue, 23 Apr 2019 18:22:01 -0000

I loved this article; going to be updating some of my models accordingly. I am curious however as to the disconnect between your approach and that taken in this paper (https://bit.ly/2Uz9geL from JP Morgan Chase, '09, on the "non-normality of market returns") which discusses the disparity between the left and right tail sizes, particularly in that the left tail is fatter than the right. In your approach above the left and right tails still seem the same even with the new distribution; thoughts on the disconnect? Or approaches which might combine the two?

Thanks again for a great article!

Re: Private: What Caused the February 5th 2018 Volatility Spike XIV termination

Vance Harwood — Sat, 20 Apr 2019 23:41:38 -0000

I've been waiting until contango gets a little stronger, 2018 was very choppy. I suggest you look at the investinvol.com solution.

Re: Backtests for Popular Long & Short Volatility Exchange Traded Products

Vance Harwood — Sat, 20 Apr 2019 23:21:36 -0000

Proshares' SVXY is still trading (and has options available) --but at -0.5X leverage of SPVXSP instead of the -1.0X it had before. VelocityShares also has EXIV which is a -1.0X vol fund based on the European version of the VIX, the VSTOXX. -- Vance

Re: Private: Using the VIX Futures Term Structure to Predict Volatility ETP Prices

Vance Harwood — Sat, 20 Apr 2019 23:18:14 -0000

Hi Todd, The position size changes also depends on the relative prices of M1 & M2. For example, if M1 was really high in price then a straight percentage split based on number of days until expiration would overweight the front month--giving a less than 30 day effective volatility duration. The calculation. detailed here:
https://sixfigureinvesting.... gives the formula used. -- Vance

Re: 2017 Ex-dividend and Pay Dates: VT, VWO, VGK, VEU, VSS, VNQI, VEA, VPL, VXUS, EDV, MGV, MGC, VPU

Vance Harwood — Sat, 20 Apr 2019 23:03:25 -0000

www.dividendinvestor.com is a good resource for looking at past dividend performance. VNQI for 2018 & 2017 has had small or skipped dividends for Q1, but total dividends for those years have been around $2.4. No data so far in 2019 to indicate that VNQI is underperforming.

Re: 2017 Ex-dividend and Pay Dates: VT, VWO, VGK, VEU, VSS, VNQI, VEA, VPL, VXUS, EDV, MGV, MGC, VPU

curious guest — Fri, 19 Apr 2019 12:58:08 -0000

Why would you be surprised if VNQI has a dividend? Is missing 1st Qtr is worrisome? Will they catch up to their supposed 9% yield over the rest of 2019? NorthernTrust and others are saying Int'l Real Estate is a good area - why isn't VNQI performing better?

Re: Private: Using the VIX Futures Term Structure to Predict Volatility ETP Prices

Todd Snyder — Sun, 14 Apr 2019 20:46:51 -0000

You stated;
"The position size of VIX Futures held by the leveraged ETPs (e.g., TVIX, UVXY, SVXY, VMIN, ZIV) changes on a daily basis based on the previous day’s percentage moves."
Are you saying the position size doesn't change equally daily? As in if there were 20 days in the cycle, each day 1/20th of M1 would be exchanged for 1/20th of M2, but instead, the size of the position "swap" varies from day to day?

Re: Private: What Caused the February 5th 2018 Volatility Spike XIV termination

crunchor — Sun, 14 Apr 2019 14:00:55 -0000

Hi Vance,

What do you invest now? Do you still invest with VIX related product? Thanks.