Use access key #2 to skip to page content.

Any sense holding on to FAZ at $28??

Recs

14

April 07, 2009 – Comments (31)

I'm sure several folks would like to know.

   I don't mean is the market ever going down, i know it is. But i've heard about the volaililty decay, and i wonder if all the volotiliy that we've had in the last two weeks is just too much for the stock, and it will never be worth the $28 again?

   What S&P strike would get us closer? 770? 

Some random guesses would be useful if no one knows.

31 Comments – Post Your Own

#1) On April 07, 2009 at 5:13 PM, anchak (99.86) wrote:

Since you are aware of volatilty decay - I have a few suggestions.....

Simpler

(1) Take the Mar 20 and Mar 30th OHLC prices of XLF and FAZ....

Your scenario is simple if XLF goes back to retest Mar 30th....in the same shape ( looks like a similar setup - ie given the retest actually happens) -

How much decay happened between those 2 points.....

it would be a decent estimate

(2) Do an excel spreadsheet ....with XLF prices. Do price path projections ( ie similar to the up/down....that's been happening of late) and reach your price point to figure out break even.

Obviously, Options/Derivatives do best in unidirectional moves - what's happening of late is a systemic destruction of option values for buy side.

Try THIS if you want to go techno deep.

Report this comment
#2) On April 07, 2009 at 5:26 PM, GenericMike (< 20) wrote:

I'm holding FAZ now with the expectation of it hitting at least $30.

Report this comment
#3) On April 07, 2009 at 5:50 PM, portefeuille (99.56) wrote:

Volatility is not the primary issue.

You might want to have a look at this discussion (the document linked in comment #10 might be of some help) and this one (I linked some maths literature there - really easy reading, nothing scary, have eat some chocolate and red wine with it!).

You should still focus more on the fact that this ETF does more or less what it claims to do:

The investment seeks to replicate, net of expenses, 300% of the inverse daily performance of the Russell 1000 Financial Services Index.

Volatility is not the primary issue.

You might want to have a look at this discussion (the document linked in comment #10 might be of some help!) and this one (I linked some maths literature there - really easy reading, nothing scary, have some chocolate and red wine with it!)

One really easy way to see what the consequences is to use a simple spreadsheet. Something like this:

column A: the natural numbers (1,2,3,...) for the days (A1 = 1, A2 = 2, ...)
column B: index (put today's value in B1 and whatever you like below that (the scenario!)
column C: calculate relative daily changes (say you are in row 2 then C2 = (B2 – B1) / B1
column D: (-3) * column C, i.e. D2 = (-3) * C2
column E: price of the ETF (put today's price in E1,  in row 2 you have E2 = E1 * (1 + D2)), ...
column F: index scaled to 100 on day 1: F2 = B2 / B$1 * 100
column G: price of the ETF scaled to 100 on day 1: G2 = E2 / E$1 * 100

Now draw a diagram where you put column A on the x-axis and columns F and G on the y-axis and play around by entering some scenarios in column B!

Column E gives you a good approximation of the price you should expect for the ETF. The "volatility issue" is a higher order correction than can have a great influence "in the long run" but before worrying about the corrections you should get a feeling of the "primary" effect.

Now to your question:

You would need to make an assumption on how the S&P and the "Russell 1000 Financial Services Index" are correlated. To come up with a "random guess" you don't need a spreadsheet of course. A glance at a chart showing both indices should suffice ...

Report this comment
#4) On April 07, 2009 at 5:53 PM, portefeuille (99.56) wrote:

sorry for that, I repost it

Report this comment
#5) On April 07, 2009 at 6:01 PM, portefeuille (99.56) wrote:

I think the document mentioned below is interesting and IS on the volatility issue.

correction of comment #3 follows:

 

 

Volatility is not the primary issue.

You might want to have a look at this discussion (the document linked in comment #10 might be of some help) and this one (I linked some maths literature there - really easy reading, nothing scary, have eat some chocolate and red wine with it).

You should still focus more on the fact that this ETF does more or less what it claims to do:

The investment seeks to replicate, net of expenses, 300% of the inverse daily performance of the Russell 1000 Financial Services Index
.

One really easy way to see what the consequences is to use a simple spreadsheet. Something like this:

column A: the natural numbers (1,2,3,...) for the days (A1 = 1, A2 = 2, ...)
column B: index (put today's value in B1 and whatever you like below that (the scenario!)
column C: calculate relative daily changes (say you are in row 2 then C2 = (B2 – B1) / B1
column D: (-3) * column C, i.e. D2 = (-3) * C2
column E: price of the ETF (put today's price in E1,  in row 2 you have E2 = E1 * (1 + D2)), ...
column F: index scaled to 100 on day 1: F2 = B2 / B$1 * 100
column G: price of the ETF scaled to 100 on day 1: G2 = E2 / E$1 * 100

Now draw a diagram where you put column A on the x-axis and columns F and G on the y-axis and play around by entering some scenarios in column B!

Column E gives you a good approximation of the price you should expect for the ETF. The "volatility issue" is a higher order correction than can have a great influence "in the long run" but before worrying about the corrections you should get a feeling of the "primary" effect.

Now to your question:

You would need to make an assumption on how the S&P and the "Russell 1000 Financial Services Index" are correlated. To come up with a "random guess" you don't need a spreadsheet of course. A glance at a chart showing both indices should suffice ...

Report this comment
#6) On April 07, 2009 at 6:13 PM, walt373 (99.75) wrote:

You might consider shorting FAS instead. Obviously that has its risks, but at least you're not guaranteed to lose money.

Report this comment
#7) On April 07, 2009 at 7:13 PM, SolarisKing (< 20) wrote:

Do i have this right?

chart

XLF on march 18th at 9.41 and FAZ at 26.35

XLF on april   6th   at 9.43 and FAZ at 16.50

 

XLF at roughly the same price and FAZ at roughly %40 less in 18 days. Does that mean with high volatility FAZ will/could lose approximately %40 in the next 18 days?

If so that would be a loss of %80 in 36 days, more or less. So XLF would have to drop approximately 1/3 of %80, or %25 or more, in the next 18 days or less to break even?

%25 of current XLF at $9 is. . . . .$2.25, target would be $6.75. Low for the year = $5.88

? right or wrong?

Report this comment
#8) On April 07, 2009 at 7:28 PM, portefeuille (99.56) wrote:

again: that is NOT a volatility issue. Try a little spreadsheet. It can be very instructive ...

Report this comment
#9) On April 07, 2009 at 7:35 PM, portefeuille (99.56) wrote:

and in the case of the FAZ the XLF is not the underlying, the "Russell 1000 Financial Services Index" is ...

Report this comment
#10) On April 07, 2009 at 7:41 PM, SolarisKing (< 20) wrote:

I don't have to understand the reason for the price wackiness, i need to know if my target is approximately right. Then i can decide when to rid myself of the FAZ.

  I TRUELY appreciate your desire to clear this up for me and point me in the right direction, but i am afraid i might spend all night reading and typing and still not understand the full issue. 

   I just need some round numbers. Do i sell in less than a week, or can i hold for 30 days. I'm beginning to think i should sell fairly fast, like this week or next. But if it's feasible for FAZ to return to it's former glory, than i can afford to hold it till the correction, and or impending doom, that i think is comming.

Report this comment
#11) On April 07, 2009 at 7:43 PM, portefeuille (99.56) wrote:

The price difference e(t_2) - e(t_1) for any leveraged ETF is not determined by the price difference u(t_2) - u(t_1) of its underlying if t_1 and t_2 are not within the same day of trading.

Report this comment
#12) On April 07, 2009 at 8:33 PM, portefeuille (99.56) wrote:

If you follow comment #5 it should take about 2 minutes to have the spreadsheet ready (maybe an extra minute or two if you have never produced a graph within a spreadsheet, but the graph is not really necessary).

I have not found a place, where you can download historical prices for the RUFIN index, but you could have a look at the BGZ and its underlying IWB (BGZ is a "direxion Bear 3x ETF" just like FAZ).

Now you can download historical prices for BGZ and IWB (click on  "Download to spreadsheet") and start your research from there).

If you combine the two spreadsheets into one you can see "what happens when" quite easily.

 

Report this comment
#13) On April 07, 2009 at 9:17 PM, anchak (99.86) wrote:

SolarisKing : portefeuille is right in the sense that FAZ deals with Rusell 1000 Financials Index linked Futures.....However, based on the proxies.....

the best ETF proxy is IYF......

Here's the comparison

IYF: 

Mar 20:  Open:32.17 Mar30 Open 32. 32.13 Practically zero change.

FAZ

Mar 20 Open 31  Mar 30: O : 23.09

Appx $8/$31 = 25% loss.

 portefeuille : I missed this exchange between you and kaskoosek... delving in infinite and exponential expansions :)

I'll try and comeback with my angle on this - you are right - but it is volatility - you just need to break it down. You are simply looking at it as a geometric expansion....which is the sum of the parts - you need to understand what drives the movement

Report this comment
#14) On April 07, 2009 at 11:49 PM, GoodVibe4Ever (< 20) wrote:

portefeuille

I knew from day one that you will give me a hell of a good time beating the heaven out of you! :) I'll see you soon when I put as much time as you do in your Caps. Great stuff! Keep up the good work and speak more often. I always enjoy your few comments.

Est-ce que tu parles francais?

GoodVibe

Thanks guys for this great conversation! I don't know why this very important blog is not up there on the top! Sigh! :(

Report this comment
#15) On April 08, 2009 at 12:04 AM, portefeuille (99.56) wrote:

my language competence is something like this:

active:

german > english >> spanish > french > italian > dutch

passive:

german > english > dutch > spanish > french > italian

 

 

Report this comment
#16) On April 08, 2009 at 12:12 AM, portefeuille (99.56) wrote:

I had french in kindergarten (just singing!) and grades 1 - 4 and 9 - 10.

I forgot to mention latin (active it would be on par with dutch and passive below italian and about 100 times better than active, which illustrates how poor my active dutch is, I guess ...).

(last nonsense post I promise/hope)

Report this comment
#17) On April 08, 2009 at 12:22 AM, GoodVibe4Ever (< 20) wrote:

Oh! German! What a tough and sharp cookie you are? I have an uncle who is German, although he wasn't born in Germany.

Gute Nacht, meine freund!

GoodVibe

Report this comment
#18) On April 08, 2009 at 12:29 AM, GoodVibe4Ever (< 20) wrote:

By the way, all of us are humans first. I don't care much about what nationality you are or what laguage do you speak and have no generalization of how group of people behave or think. I just read my post and it didn't sound right to me. Speak of odd stuff! :)

By the way, I will enjoy some of this nonesense if you tell us who are you (if you like and have the time). 

Take care! Nice meeting you.

GoodVibe

Report this comment
#19) On April 08, 2009 at 1:35 AM, RNP1971 (< 20) wrote:

Basing an important decision on a hypothesis formulated around and tested by actual past data leaves you open to at least a systemic congruency bias. 

There is a much better chance of developing a useful model focusing on probable possibilities instead of history. While  historical data is referenced when considering probability, these are very different and  a clear distinction should to be made.

In the case of FAZ, it has been trading for a short period of time there are plenty of known unknowns, and (of course) untold numbers of unknown unknowns.

 

I suspect nothing I have said is novel to Portefeuille who presents a valid process for viewing various possibilitites, ,but I believe emphasis needs to be added to his discription of the spreadsheet model as

>>>'what happens when'<<<.

This is great for getting a comfort level with the expected magnitude of movement in relation to an index, but this definately should not be extended to a predictive context for actual prices.  Portefeuille, please correct me if I misunderstand your intent.

 

In short, I suggest being wary of basing a decision relating to probability of future price ranges, that uses historical data in favor of analyzing probable possibilities.

 

It is late and I don't think that i have conveyed the idea very well.  So if this is unclear, but it sounds like I am making reference to something with which you are completely unfamiliar,  take a moment to Google at least:

congruency bias,

conjunction fallacies,

availability heuristics,

and patternicity.

Report this comment
#20) On April 08, 2009 at 2:57 AM, portefeuille (99.56) wrote:

By "you can see 'what happens when'" in #12 I just meant that you could study what effect changes in the price of the underlying over some period of time have on the price of the ETF over the same period. I did not suggest predicting the future price development of the underlying (and thus of the ETF) from historical data.

Report this comment
#21) On April 08, 2009 at 8:30 AM, anchak (99.86) wrote:

Folks.....If you want to really get it right - you have to go the simulatory way ( ie scenario running) that  portefeuille and others suggested.

Ie generate some possible price paths for IYF and then do the    -3x computation and compound it to see what it would do to FAZ.

Solaris...my suggestion is more of back-of-envelope because looked like that's what you were after for an initial blush.

portefeuille: What I was alluding to in the above post is the mix by which Direxion is maintaining the 3x leverage.

They actually hold the underlying ( I was surprised by it - Proshares ie SKF goes the swap way) - to a 25% allocation.

And then use the 75% allocated to the Futures and Cash to do the leverage piece.

Essentially

(futures+ cash) = 3*(Underlying)

But since the partial derivatives (futures) is not aligned 1 to 1 with Underlying - they essentially adjust/continuously rebalance the cash to futures allocation...and its due to this fact why volatility comes into play...because the futures are going off whack...

due to an interplay between underlying+time + volatility.

Hope that made sense

 

Report this comment
#22) On April 08, 2009 at 1:27 PM, portefeuille (99.56) wrote:

There has been written quite a bit about these leverages (inverse) ETFs. I thought about writing some kind of summary of what I have found but maybe a link collection serves better since this is hardly my "home turf". So here it is in a very raw form (just something to click through on a rainy afternoon): 

1,2,3,4,5,6,7(!),a,b,c,d,e,f,g(FAZ),h(FAZ),8,9(2 videos at the bottom),10 (I suspect the term "volatility decay" got some impetus from this article),11,12,13,14(FAZ),15(!),16,17,18,19,20(calculation at the bottom), ...

and unrelated but nice: this

Report this comment
#23) On April 08, 2009 at 2:25 PM, portefeuille (99.56) wrote:

Having skipped through some of these articles and the above discussion again I think I should clarify that when I said "Volatility is not the primary issue" I was referring to "tracking errors" and NOT (!!!) the effect of ups and downs in the price of the underlying on the ETF over longer periods of time.

But I think this (the compounding and "not delivering the 2x return over longer periods") is what many have in mind when talking about "volatility (decay)". As of yesterday I was not aware of the article 10 so the confusion is probably more due to a terminology mismatch than a "fundamental" disagreement. I think when anchak wrote #20 he had already noticed that.

Someone should write a little summary of all this "leveraged ETF" stuff (including some of the "maths" and "derivatives" aspects, maybe someone could verify whether this rebalancing has really become such a dominant factor in the "last trading hour" as some in the abovementioned articles suspect (I found link 3 here) ...

Report this comment
#24) On April 08, 2009 at 2:33 PM, mistermiranga (98.04) wrote:

Good stuff here, thanks guys. The leveraged heat is very tempting and I admit to using it irresponsibly. The thinly traded stuff seems particularly nasty when things get sideways...

Here is a post I saw this morning on Seeking Alpha on "rebalancing" and holding the heat long term.

 

Report this comment
#25) On April 08, 2009 at 6:02 PM, portefeuille (99.56) wrote:

An alternative to leveraged ETFs are "Knock-out products". They have the "Knock-out" effect and are by no means "equivalent" to leveraged ETFs, but they are very good at realising a fixed "return-multiple" over longer time periods (up to several years) and you have thousands to choose from (lots of underlyings and quite a variety of products for most underlyings). You can read about them here (click "Knock-out products" then on "Knock-out products basics" or start with "Quick search"). One example here:

You can see that it does track the underlying quite nicely ("qualitatively").

The price is calculated in the following way: p = (u - s) * m * c, where

p = price of underlying (in this case the DAX index in EUR (1 point = 1 EUR)

s = strike price (here: 2358.177)

m = multiplier (here: 0.01)

c = currency conversion factor (nothing to convert here since the DAX is "calculated in EUR" as is the price of the knock-out certificate, so here you have c = 1).

The Dax closed at 4357.92 (=: u1) at the close of Xetra trading today (=: t1) (11:30 a.m. ET, 5:30 p.m. (= 17:30 !) local time.

Thus you get p = (4357.92 - 2358.177) * 0.01 = ca. 19.997 EUR = ca. 20.00 EUR (a round number by coincidence only!).

As you can see in the intraday chart the price of the underlying was indeed close to 20.00 EUR at 17:30.

Now where does the leverage come from. Suppose you had bought the certificate at the close of Xetra today (:= t1) for p(t1) = 20.00 EUR and hold on to it until time t2. Say the DAX is then at u(t2) points. Using the formula for the price you get

p(t2) =  (u(t2) - s) * m     (F1)

Now the relative change of the underlying (the DAX) over the period (t1,t2) is

diff_DAX = (u(t2) -  u(t1)) / u(t1)

and the relative change of the knock-out certificate over the same period is 

diff_certificate = (p(t2) -  p(t1)) / p(t1) = (u(t2) - u(t1)) / (u(t1) - s).

This gives you the "leverage" factor l:

l =  diff_certificate / diff_DAX = u(t1) / (u(t1) - s)     (F2)

So the leverage is indeed constant (no dependence on t2!) if the formula (F1) holds over time. (F1) does hold quite well for some of the certificates, but you have to study the different kinds of certificates a little bit to know how they work (they have different "mechanisms" and some are NOT meant to have constant leverage, but I will not get into that, I just wanted to show that you have more than the leveraged ETFs to choose from - have a look if you like!).

Now one last glance at (F2). You can easily see that the closer the underlying is to the strike price the larger the leverage (AND of course the higher the probability of future knock-out (you cannot ask for too much (!!!)) so by choosing s (there are usually certificates with different strike prices for a certain underlying) you can choose the leverage (and risk of knock-out event) you like and with these certificates you cannot lose more than 100% in contrast to buying "on margin" like you do when using futures or CDFs (you can of course lose more than 100% if you buy the certificates "on margin").

That's it!

Report this comment
#26) On April 08, 2009 at 6:14 PM, portefeuille (99.56) wrote:

and sorry solarisking for "taking over" this post!

maybe a "don't know" or silence on my side would have better served you. I am really not sure about the FAZ (anything can happen).

Report this comment
#27) On April 08, 2009 at 8:02 PM, anchak (99.86) wrote:

 portefeuille:

Excellent ! You are a great asset to CAPS man! I was not aware of these - I am curious how many of these are available to retail customers.....and of course whether I can buy these from a US broker ( I guess I'll have to check with Interactive Brokers).....

I think you are fundamentally correct ....especially with 1x and 2x leveraged ETFs - which are equivalent to a Deep- in -the-money option. These products by virtue of providing LONG-TERM constant leverage  [ As by your math you have established that's it dependent on time period zero - you used 1 ...but I thought zero base is more appropriate :) ] - are an excellent alternative. The biggest issue is the knock-off trigger - if the underlying goes below strike - you are have a 100% loss.

My guess would be - that's where the premium would come from - ie an index linked option on the Dax at a similar strike would have a differential starting Leverage and return potential than this product - that's how the issuer is getting compensated.

But I feel you are right about these being a better product for "investor" types who are delving into leveraged products.

Any guesses on expense premiums etc? Or is it baked into the price?

 

Report this comment
#28) On April 08, 2009 at 8:58 PM, portefeuille (99.56) wrote:

I will look into that (they do get the bid-ask-spread). All of them are available to German retail investors via the exchanges in Stuttgart (EUWAX) and Frankfurt (scoach), I don't know about US citizens (but I am pretty sure they can trade them too). I have no idea why I did not mention them earlier. The certificate (if it is a "call", I forgot to mention there are puts as well, but it is pretty obvious how they work, when you know how the calls work ...) stops trading before maturity (some are without maturity) when the price of the underlying drops below the "Knock-out barrier" (didn't mention that one either). Usually strike and knock-out barrier coincide at the beginning (first day of trading of the certificate) but the knock-out barrier is usually raised from time to time (you would have to look that up "in the small print", I guess). If the knock-out barrier is higher than the strike (again, I am talking about the "call" version) when the underlying goes below the knock-out barrier then the certificate has a non-zero final value (simply determined by p = (u - s) * m, where now u is the knock-out barrier, unless the issuer has not been able to "get out of his 'hedge' positions" without further losses in the time following the knock-out event (for example after a large gap down or when it keeps falling very fast). So you may not lose 100% when the underlying falls below the knock-out barrier).

And you can, of course, buy a new certificate with lower knock-out barrier and strike when the underlying "comes close to" (or has fallen below) the knock-out barrier. You could start right away with several certificates that have different strikes but the same underlying or you can readjust your risk by selling your certificate and buying one with a higher strike (of course you do change your leverage by doing that) ... sky is the limit!

Almost everything I have written above about these certificates was "off the top of my head" but should be more or less accurate.

Report this comment
#29) On April 11, 2009 at 10:57 PM, herztical (28.31) wrote:

...there are so many better trading strategies to use rather then using this garbage (FAZ/FAS) and for that matter any levered fund. Yes, I have them in CAPS bc it's quick points and no real $$  at stake. I also read somewhere that the average holding time is 13 minutes for these stocks. What a great investment!  Black jack hands take longer!

Short financials = yesterdays trade...been saying that for a while now.  Move on folks, there are much better shorts out their or pair trade opportunities

Report this comment
#30) On May 31, 2009 at 6:53 AM, kaskoosek (92.46) wrote:

This is crazy

Report this comment
#31) On August 07, 2009 at 12:55 AM, ChrisGraley (30.25) wrote:

I missed this post before, but Porty just linked to it from another blog.

For what it's worth, it's a simple math problem.

You simply have to figure out the price that you are hoping to sell it for.

Lets say that you want to sell it for $42.00

That would mean that your hoping for a 50% increase in the fund. That may seem impossible in this given market, until you remember how the fund is designed. Time is on your side in prolonged drop in confidence of the financial sector as well as the leverage. Yes you will see long term decay, but that is irrelevant if you're worried about whether you should sell now.

If the Russell Financial Index were to drop 17% fairly quickly, you should see an increase pretty close to 50% in FAZ remember FAZ is designed to go up 3% for a 1% drop in the index.

Now lets say that the index drops slowly, but for a long period of time...

Lets say it drops 1% every day for 2 weeks

First Day $28.84

Second Day $29.7052

Third Day $30.596396

Fourth Day $31.51424668

Fifth Day $32.4596740804

Sixth Day $33.433464302812

Seventh Day $34.43646823189636

Eighth Day $35.4695622788532508

Ninth Day $36.533649147218848324

Tenth Day $37.62965862163541377372

Eleventh Day $38.7585483802844761869316

Twelfth Day $39.921304831693010472539548

So the index can go up 17% in one shot or only 1% in a little over 12 days to gain the same 50% in FAZ.

This is not figuring decay or market reaction though, and the compounding of the interest doesn't seem like a big deal given that I used a slow drop of 1%, but look at the case of a drop of say 8% in one day followed by a few days of 2 or 3% and you can see where compounding is important.

Once you know what you want to sell it for, it's just simple math to figure out what the financial sector has to do to make it happen. 

Once you have that picture, it will make your decision easier.

I hope you're talking about CAPS btw, because FAZ was never meant to be more than a trading vehicle in real life.

I hope that helps.

Chris

 

 

Report this comment

Featured Broker Partners


Advertisement