Friday, September 01, 2006

SPY "noise" trading?

The "random walk" of the market, is just another way to say "noisy" walk. Daily increase and decrease of prices usually trace out a path that is very similar to the volume of the hiss that you get from a radio tuned to a channel with no stations. One difference is that markets also have "drift" -- they tend to go up or down at a longer time scales based on macro economics, wars, fear & greed, and who knows what else. For reasons that are not clear to me, the magnitude of the "noise" on a market tends to be proportional to the square root of the time span. For example if the magnitude of the annual noise is 12%, the daily noise will tend to be around 12% divided by the square root of the number of trading days in the year -- about 250. The square root of 250 is about 15.8, so the daily change in this example would be 12%/15.8 or around %0.76 The minute to minute variations are predicted to be 12%/346 (square root of 250*8*6) = ~ .035%. On the Spy S&P 500 index this would be around 4 to 5 cents. Eyeballing the charts for daily and by minute high vs. low these predictions look pretty close.

So is there a way to take advantage of this level of predicted noise, without having to guess which way the market will "drift"? I think not, because if people had figured this out they would be writing books about it while sitting on their decks in Maui.

The big challenge is to avoid getting caught if the market drifts in a direction contrary to your position. Let's say you buy SPY at opening at $130 with a sell order set at $130.99 (0.76% higher). The theory would predict that you would succeed with this strategy about 50% of the time on the 1st day. If the market is "going sideways" with no clear overall trend or trending up, the odds are very good that you would close out your order successfully within a few days. However, if the market decides to tank at that point you may have to wait a very long time for your sell order to execute. So is there a cost effective way to collect the likely day-to-day gains but not get wiped out if the market tanks? One approach would be to buy puts at the current support level of the trading range. If the market tanks you have at least limited the potential damage. Your loss would be the difference between your buy-in price and the put strike price and the put option premium. Another approach would be to hedge with $VIX volatility options which tend to go up when the market goes down. These have the advantage that they are profitable if the market drops suddenly, regardless of what prices the SPY index is selling for.

Data on $VIX options (Volatility) and interest rate options

  • $VIX Expiration Date:
    In most months VIX options expire the Wednesday before equity options expire. Some months however they expire on the Wednesday after the 3rd Friday of the month -- which is generally when equity options expire)


Reference http://www.cboe.com/Products/indexopts/vixoptions_spec.aspx


Interest Rate Options Product Specifications

Symbols:
13-Week Treasury Bill - IRX
5-Year Treasury Note - FVX
10-Year Treasury Note - TNX
30-Year Treasury Bond - TYX

Underlying:
IRX is based on the discount rate of the most recently auctioned 13-week U.S. Treasury Bill. The new T-bill is substituted weekly on the trading day following its auction, usually a Monday. FVX, TNX and TYX are based on 10 times the yield-to-maturity on the most recently auctioned 5-year Treasury note, 10-year Treasury note and 30-year Treasury bond, respectively. Options are European style exercise.

Dividend Trading -- what I learned this summer

What I learned this summer is that my dividend trading approach (buying stocks right before ex-dividend, and hedging by selling relatively deep in the money calls) works most of the time. Unfortunately it does not successfully handle the negative side of the "long tail" of market volatility. The long tail refers to the market's tendency to become much more volatile than normal at times--more often than the standard "normal" distribution would predict. I was caught with large positions in several stocks where the options were not assigned at the ex-dividend date (which is normally a good thing because I obtain the dividend with a good call option hedge), but then the overall market went into a nose-dive that ate up my 5% to 10% buffer on the stocks and promised significant loses if the stock continued to tank. I bailed out in both cases before my stock went out-of-the-money on the option, but I still took losses that will take a long time to recoup with the standard dividend trading sorts of returns. In both cases I would have been alright if I had held on, but one of my primary goals with dividend trading is to avoid the necessity to predict the direction a stock is going to go. This stress is something I don't need. Watching 2300 shares of the stock go up and down several points day (FCX) near my option strike price is not good for my blood pressure.

As always the market is very good at preventing free lunches and it requires significant risk taking to even have a reasonably priced lunch. I have spent some time looking for ways to avoid this risk, including stock futures, but so far I have not found anything that works to protect against the negative long tails without reducing the rate of return to CD levels. The opportunities that are still worth a look are the ones where the stock option expiration is only a few days after the ex-dividend date. Unless really dramatic things happen then the risk looks reasonable. The problem with these situations is that the premium available on the option is low, and the option is more likely to be assigned because there the premium available is low. The other thing I want to look at is whether the gains from the short option in a market sell off could be use to finance enough futures contracts to hedge the overall position successfully without killing the rate of return. One fairly promising approach would be to buy $VIX calls to hedge the negative long tail scenario. A quick look at the 2300 share FCX situation I held in July indicates that just a couple of $VIX $12.5 calls would have protected the position.

Monday, August 14, 2006

Limit orders vs Market orders -- selling / buying "at the market"

First of all, I rarely use market orders even when I just want to get the market price. I use limit orders where I specify the minimum price I will accept to buy or sell on my order. A broker's software will always figure out which is better for you (e.g. for buying a stock the limit or lower is better). The only time I use a market order is when the stock has a minimal bid/ask spread (a few cents at most) and is behaving in an orderly fashion or if a stock is moving very fast and I don't want to be left behind (not recommended). I use limit orders because I have seen market orders execute at prices that don't bear much resemblance to what the real time quotes seem to indicate. Occasionally the price was better than I expected, but usually it is worse. I think this might be due to the multi-tiered nature of markets. For example your broker might try to match your order with the opposite order another one of their clients has just entered (e.g. buying what you are selling) . They love this because then they get to keep all the commission money--but it takes time. After being burned a few times I learned my lesson and I don't take the risk of market orders. With options it seems like the market prices often change adversely as soon as you enter a market order. Options markets are so "thin", so few buyers / sellers relative to stocks, that I never use market orders. If I enter the limit order at the most recent execution price or a little less favorable to me there may be a few minutes of angst waiting for the trade to go, but the orders almost always go through. The exception is when the stock or option is making a big move you may end up "chasing" an stock up or down--not a fun situation. Even in those situations there are usually minute-by-minute ups and downs that will let you get in or out if you don't get greedy. The key is to remember that you have made the decision to buy or sell. Usually those few cents one way or another don't impact your overall profit or loss significantly. For orders where you are doing combinations of things (e.g. buying a stock and selling covered calls against it), then you should use "debit/credit" orders that specify the net and render the exact execution prices unimportant to you. Not only is this a much less stressful process, it often saves you money through better prices and lower commissions. See my post on these.

Friday, June 30, 2006

The ins and outs of debits & credits

Notes on debit & credit orders.

Tutorial: Debit and credit orders are combination orders used to simultaneously buy or sell different security types (e.g stocks and options), where you don't care about the prices you pay/receive on the individual securities, just so long as the total is equal or better for you than the amount you specify.

These are useful for several reasons:
  1. You remove the risk that the stock will move against you between the time you put the 1st position in place until you can get the second in place

  2. Often you drop into some internal sanctum of brokers where the pricing rules that mere mortals have to adhere to are missing. For mortals you can't go below 1 cent increments on your stock price, or 0.05 for options below $3, and 0.10 on option prices above $3. With a debit or credit order you will sometimes see your stock price go to 3 decimals (e.g. 127.238) and the options down to the penny

  3. With these orders you are in a much better position to try and beat the "spread" between the bid (what the market maker will buy your stuff for) and the ask, which is what the market maker will sell the security to you for. When the stock is jumping around on a busy day it is scary to make an offer between the bid/ask because you might miss out if the security moves the wrong way. Even tougher, if the bid / ask on an option is only a tenth for a option above $3 (e.g. 4.1 bid / 4.2 ask), you can't even submit an order that splits the spread.

    To keep debit and credit straight, my mnemonic is : debit -- me going into debt, credit -- me getting the credit. I use debit orders for doing a buy/write (buying stock and selling covered calls) or credit orders for closing them out. Don't try to use "market" combination orders, these aren't handled automatically in some cases and can take a significantly longer period of time for one half to execute -- which defeats the whole purpose of the debit/credit order.
  • I had several recent debit orders execute quickly -- which implies I was leaving money on the table (or in one case I was just being stupid). The coarseness of the option pricing vs the underlying typically adds 0.1 of slop in the pricing (on options above $3 anyway)--so that needs to be factored into your bid plus at least splitting the bid-asked spread.

    It isn't hard to change the order, so I'm thinking I should start at 2/3rds of the spread + the slop factor should be the starting point. For example, for a buy/write: if the underlying is at $54.35, the $47.5 calls at 6.90 bid, 7.10 asked (0.05 of premium on the bid price) . Then, the slop is .05 at this point (the option probably won't go to 7.o bid, 7.2 unless the stock reaches $54.40 ). The spread is 0.2, so 2/3rds of that is 0.13. So a reasonable starting debit offer for a buy/write would be: 47.5-(0.05 premium already in bid price+ 0.05 slop + 0.13 bid/ask split) = $47.27. This is an offer designed not to leave much money on the table, but not wasting your time with a no-go offer.

  • In some cases debit /credit orders don't work well at all. If you order is split up and each part goes to a different exchange, then the tight time coordination and ability to beat the spread is lost. If you see stock / option pricing go by that should have triggered your order, but didn't, this is probably what happened. You should cancel and drop back to manual control at that point.

A little further down the path on dividend trading--late June

Regarding option assignment, I obtained a fair amount of data this week.
  • Out of the possible 17 options contracts I was short on that could have been exercised for dividend capture only one, a BestBuy (BBY) option was assigned. This data suggests that if you can lock-in a little profit in the few days before a stock goes ex-dividend on a buy/write that the odds of being assigned are significantly lowered--even if the option is significantly in the money ($10 in the case of BBY)
  • One other data point -- the fact that only one of my 7 BBY options got assigned gives me hope that playing the odds on the Strike prices that only get partially cleaned out is a viable strategy.
  • In the school of hard-knocks category, I closed out my GSF position, but adjusted the credit offer the wrong way--I was't thinking this was the point at which I wanted to increase my offer as much as possible--I guess I have been doing too much buying and not enough selling.
  • That's the bad news on GSF , the good news is that either with the assigned call (profit of .21 per share, or with the dividend plus option premium (profit of 0.425 per share) this was a strategy yielding 12%+ on an annualized basis.
  • Final analysis on IVV (SPY wannabe) options -- none of the options were exercised at ex-dividend time, plus finding out the ex-dividend date is a bear. Skip this next time around.
  • Final analysis on FBR. 6.8% annualized yield ex-dividend 28-June. I tried to get into $7.5 options buy/write without success (not even a 7.5 debit worked). $10 options on the 27th might have been a fair play. The stock was at 11.5 and about 50% of the $10 options did not get exercised. The stock was very volatile however. All of the $7.5 options were exercised--so the $7.5 buy/write was a loser strategy from the get-go with no lock-in profit available.

Tuesday, June 27, 2006

Observations on Dividend trading -- buy/writes

I will know more in the next day or two whether this stuff works the way I expect (options assigned), but I have already learned quite a bit with my recent intensive trading.

A quick tutorial. I am trying to make a small profit by buying a stock and simultaneously selling in-the-money options against the stock. I do this at a price point that locks-in a profit --where the price I pay (called the debit) is less than the strike price of the option. For example, if the stock is at $38 and I write $30 calls I will pay less than $30 for the combination (perhaps $29.80). My lock-in profit is this case is 0.20 per share. You need a fair amount of captial to make this worthwhile... I try to sell options that are well in the money (high intrinsic worth) because this protects my investment (Delta approx. -1) and increases the odds that the option will be assigned when the stock goes ex-dividend. If the option is assigned my position is automatically closed out and I immediate get lock-in profit (minus commisions of course). If the option is not assigned, I get the dividend from the stock, but the option price jumps up in the short term to negate that benefit. You have to wait (and tie up your money) until the option expires to get your full profit out in that case.

Some notes:
  1. It's you against the computers. The market makers seem to make a decision whether they are going to allow any profit lock-in (buy/write with debit less that option strike price) on a stock by stock basis. On very high dividend stocks this is a given. When you get into the stocks yielding around 2% per year it looks less consistent.
  2. If the bid value is near or above the breakeven value you are good to go. You can hope to split the spread or better. The longest I have seen a successful trade take is about an hour. Most of the time it goes pretty quick--making you feel like you have left money on the table.
  3. So far I have seen the buy/write stuff work well when the market makers are willing to play. Otherwise you better be willing to take a loss up front and hope your options don't get assigned. I'll be looking closely that that assignment data in the next couple of days.
  4. The big open issue right now is whether the options where the market makers are playing are going to be assigned. If not I will see the option volatility jump up and it could take until the options expire to get out with my profit (but it will be the buy/write profit plus the dividend)

Sunday, June 18, 2006

Late June Dividend events & notes

Stock Ex-Div_Date Div Recent_Price Div/Recent_Price%
FBR __ 28-June ___ .2 __ 10.11 ___ 1.97%
SPY __ 16-June ___ .555 _ 124.65 __ 0.45%
IVV __ 23-June _ .528 __ 125 ____ 0.45%
BBY __ 29-June ___ .08 ___ 52.5 __ 0.15%
DE ___ 28-Jun ___ .39 ___ 80 ____ .48%
GSF __ 28-June __ .225 ___ 53 ____ .42%
CCU __ 28-June __ .187 ___ 30 ____ .62%
DOW__ 28-June __ .375 ___ 38 ____ .98%

Notes
  • Tried without success to get any bites on buy/writes on DOW with debit amounts giving me any profit at all. My CCU order barfed on Schwab when I tried to do the buy/write at $25
  • GSF on the other hand happily gave me a profit of 0.21 on 47.5 calls, I had tried 0.25 for a while with no takers, but 0.21 went immediately--left a few pennies at the table I guess. Ex-Div is tomorrow, so I expect/hope my position to be short-lived.
  • I tried a zero profit offer with DOW -- ( $35 call with a debit of $35, but no takers). I ran a "science experiment" with DOW to see if the buy/write was not working well. I bought DOW at 38.01 and then tried to split the 3.0 / 3.2 spread on the $35 call with a 3.1 offer. Even though DOW got up to 38.04 there were no takers. I chickened out at that point and took my $28 in profits....
  • On FBR: I tried a Buy/Write for $7.5 July Calls at 7.45 debit -- no takers
  • On FBR: Since the options will almost certainly be assigned I wanted a little money up front
  • On BBY: I did a Buy/Write for $42.5 July Calls at 42.35 debit -- it took an hour to fill!
  • On BBY: It will interesting to see if the BBY Calls get assigned -- 3 weeks left before expiration
  • Found an interesting web site www.amextrader.com -- gives ex-dividend dates a few days in advance

Thursday, January 26, 2006

Capturing dividends--the path to low risk returns?

For a long time I have been intrigued with instantaneous jumps or in trader parlance "gaps" in stock prices. As near as I can tell there are 3 sources of these sorts of jumps where a stock price makes no attempt to trade at intermediate values, but rather pops up or falls immediately to a new level:
  1. News surprises (e.g. due to earnings warnings, analyst upgrades, business deals, market panics),
  2. Large blocks of stocks being bought or sold, and
  3. Dividends. Dividends are interesting because they are announced ahead of time to happen on a certain date. On the "ex-dividend" date any shareholder that owned the stock prior to that day becomes eligible to receive a stated amount of money per share of stock that they own. It doesn't matter if you have owned the stock for one day before or 5 years --that money shows up in your brokerage account typically a couple of weeks later on the "distribution date".
The first two are inherently unpredictable, but wiht dividends, at least the date, and usually the amount is predictable.

With dividends the amount of money can be significant--for example NAT just paid out a dividend of $1.88 on a stock that was trading around $36 -- a 5.2% payout. At 1st glance this looks like a great deal, but like all things on Wall Street there is no free lunch. For starters, at opening, the day that the dividend is locked in--the ex-dividend date-- the stock typically drops an amount that is equivalent to the dividend. So, if you buy the stock to collect the dividend, you need to remember that the stock+dividend price at opening on the ex-dividend day will give you about a wash. The fun then begins as to whether the stock bounces back up to the previous day's level or drops due to participation in the general market action. Welcome to day trading....

Because of this dependence on market action, just buying stocks to capture the dividends is an iffy deal. I will write more posts on this subject detailing some of my attempts at capturing dividends without being immediately embroiled in market action.

Friday, January 20, 2006

Hobo's Stock list

GoGo Stocks
*AAPL (Apple) IPOD rules, Mac fades --is it just the transition to Intel, or is Apple a has-been in the computer business? Latest earning really looked pretty good--however their forecast is weak
*SBUX (Starbucks)

High Tech
*ALTR (Altera) FPGA mfg
*XLNX (Xilinx) FPGA mfg -Missed their Jan06 numbers after raising their guidance twice--CFO looks like a fool. Stock hammered 8% today (20-Jan-06)
*SNDK (Sandisk)
*QCOM (Qualcomm)
*RMBS (Rambus)

Retail
*AMZN (Amazon)
*BBY (Best Buy)

Old Guard High tech
*CSCO (Cisco)
*DELL (Dell) Business model in trouble, can they pull it out?
*HPQ (HP)
*INTC
*A (Agilent)

Internet
*EBAY
*GOOG Still not in the S&P 500 for a while, Market cap is huge
*YHOO Jan06 report disappointing

Big Dividends
*HSVLY

S&P 500 Index funds
*IVV (ex-dividend a few days after SPY), exploit the differences from SPY?
*SPY (ex-dividend 3rd Friday Dec, Mar, June, Sept)

Schadenfreuden stocks (look it up)
*MSO (Martha Stewart)
*SUNW (Sun Microsystems)

RAMBUS -- What goes up must go down? Or maybe up some more...

RAMBUS (RMBS) has merely doubled in the last month--on nothing more than news. A design win here (AMD) , a favorable lawsuit ruling there, an analyst's upgrade with a $936M in revenue upside ($157M actual in 2005). The earnings announcement yesterday was a yawner-- 9 cents a share instead of the one analyst's estimate of 8 cents. The stock did nothing on that news. So with a PE ratio of 117 (compared to Google's stogy 96) what should we look for on this stock?
I'm betting that it will not just stay stuck at 34 like it is right now. I sold 1000 shr short (at 34.43) and bought 12 Feb $25 Calls (at $1043 apiece). The delta right now on this combination is 1 -- at this price any change in the price of the stock will cause offsetting changes in the shorted stock and the call. However--if there is a reasonably big move (a couple of points) things get interesting. On the upside the delta on the calls will move from .89 towards 1 and the position moves into the black. On the downside the value of the options drop, but at a decellerating rate as the delta of the options drops. The profit /loss Shorted stock of course moves dollar per dollar with the stock. Right now I am wondering if I should set up a standing order with my account to close out the position if the stock moves sharply in one day (as it did the day I opened this position). One other comment, the volatility of the call options dropped significantly after the earning report--from around 100 to 90. Not surprisingly since there is less uncertainty about the stock. This might be a behaviour to exploit-if you are short on the options, not long like I am now.