Option Database Explorer User s Guide

Transcription

1 Option Database Explorer User s Guide

2

3 Table of Contents 1. Introduction Installation... 5 How to Install... 5 LicenseKEY and Purchased Database Installation How it Works General Aspects Rule-Based Trading System Implementation Cadence, Symbol and Dates Strike Selection Backtest Box Exit Logics Time Exits Profit Target and Stop Loss Automation and Defaults Option Language Defaults The Option Chain Window Price Modelling and Database Adjusting The Performance Report Grouping Trades TS Code Generator for Naked Shorts Hedging Hedging Strategy Code Launching the strategy Confronting Results Exporting Results Miscellaneous Test Name Save and Open a Backtest Clear All Data Folder

4 1. Introduction The Option Database Explorer is a software aimed at testing and evaluating trading systems based on options on futures and their underlying instruments. We provide our users with an options on futures price database based on the symbols needed by the subscriber. Because of the intrinsic complexity of the task, option price databases are not easy to navigate. This is partially due to the massive number of contracts available for each underlying instrument and partially to the nature of the variables involved. When navigating a futures price series, the main complexity is to select an appropriate algorithm for contracts aggregation that allows a multiple instruments database to be evaluated as it was a single time series. When dealing with an option database this is not possible, the multi-dimensional nature of the task makes reducing intrinsic complexity not a feasible way forward with regard to the problem of data navigation. Our approach has been to create a framework that would allow the user to navigate an option price database with ease, hiding the underlying complexity behind a friendly user interface. In this guide, you will find instructions on how to use the Option Database Explorer for testing and evaluating a trading system. For a detailed list of the markets available, please refer to 4

5 2. Installation How to Install To install the Option Database Explorer you have to download the installer first, request it at Execute the installer and follow the instructions. When choosing the installation directory consider that by default it will also be the location of the database. If you prefer you will be able to choose a custom directory in a different location for your database, but this will require extra steps. 5

6 Press Next two times and wait for windows to ask for the authorisation to install. Wait for the process to be completed, it may require a few seconds. After loading has finished press Close and installation will be completed. At this point the program is not ready yet, in fact, in order to function properly, the database has to be extracted. Go to the installation folder (default: C:\Program Files (x86)\optionlab\optiondb Explorer). Inside you will find an archive (.rar file) called Database (DEMO).rar Right-click on it and press Extract Here as shown in the picture. Wait for the database to be extracted, at the end of the process a new folder called Database (DEMO) will automatically be created. At this point you can delete the original.rar file. (The folder containing the database MUST be named Database (DEMO), otherwise it won t be found by the software). Inside the folder just created is stored the demo sample of the database. 6

7 The Option Database Explorer is now ready to be used. At this point you can execute the program, and after a brief check on internet connection the following window will appear: Press Continue to proceed, then, after reading the Disclaimer, press OK. You are now ready to use the Option Database Explorer, try the software for free. LicenseKEY and Purchased Database Installation If you purchased the Option Database Explorer you have received a file called Option Explorer License KEY and Database.rar. Inside you will find a folder called Database and a file called License KEY.txt. It s important to point out that both the License Key and the Database will be linked to the machine of first activation, thus we suggest activating the Option Database Explorer on the computer on which you are willing to use it. In order to be activated, the Option DBE has to be launched. If it has not been activated yet, after a brief check for internet connection status, a window asking for the license key will be shown. 7

8 Here you can find information about your demo subscription expiration date and controls for PC registration of you license, you will find your license in the file License KEY.txt. Enter your License KEY and press Activate. Your computer will be registered and a confirmation message will inform you that the registration process was successful. Next step will be to install your new Database. In order to do it you will have to move the folder database contained in the file Option Explorer License KEY and Database.rar in the installation folder of the Option Database Explorer. The default directory is C:\Program Files (x86)\optionlab\optiondb Explorer, but if during installation a custom location was selected the database will have to be moved to the appropriate directory. 8

9 Even if it is not necessary to delete the folder Database (DEMO) we suggest doing it because the DEMO sample cannot be accessed by the full version of the Option Database Explorer, and keeping it would result in a waste of disk space. Once the Database moving process has been completed the installation folder should look as follow: The folder that contains the database MUST be named Database, changing its name would require manual indication to the software on where to find it. The folders contained in the database MUST not be renamed, doing it would result in the program not functioning properly. If you want to put the database in a custom location you should follow these steps. Move the database folder in your directory of choice. In the figure the selected directory is D:\Cartella di Esempio. If you have chosen the default directory for your database, it will be automatically found at program start-up, and a confirmation message in the Log dialog will inform you of the success as shown in the figure. If you have chosen a custom location, when launching the program for the first time, the following error message will be reported: This happens because the database is not in the default location, and a custom directory has never been specified. 9

10 To manually set the database directory click on Edit -> Select Database Location. Provide the program with the directory of the folder Database, selecting it in the dialog box as shown, then press OK. The Log will inform you that the operation succeeded. As long as you don t move the database folder, this operation will not have to be repeated, the Option Database Explorer will remember your custom location at any future start-ups. Your Option Database Explorer full version is now ready to be used. 10

11 3. How it Works General Aspects The purpose of the Option Database Explorer is to allow the user to test trading systems that are based on options on futures. The program provides the user with easily accessible high quality option historical information. Available data, for every available contract at any given point of time (on a daily basis) features: - Price - Delta - Gamma - Vega - Theta - Rho - Implied Volatility - Volume - Open Interest To test a strategy means to simulate the rule-based orders generation process on past data, to see how it would have performed during the time span of the backtest. This means that a series of historical trades has to be generated before some valuable information can be obtained. To generate an option trade some information must be provided. First, the opening date of the trade to screen the contracts that were available at that time from those who weren t. Entry (and exit) dates are enough to identify a trade in Futures/Stock backtesting, but not when working with Options. Further information has to be provided in order to select a contract from the multitude available. An option contract is identified by three coordinates : - Maturity Date - Exercise (Strike) Price - Option Right: Put/Call Once these three fields are known, along with dates, an historical trade can be generated. Rule-Based Trading The selection of these 1+3 coordinates (Dates + Maturity/Strike/Right) should be based on some kind of algorithmic logic. At this time, manual implementation of the orders generation algorithm is the only option available. This means that the calculation of the 1+3 coordinates will have to be performed by some external tool and then manually entered in the Option DBE, or be based on some easily reproducible set of rules so to make direct manual selection possible. 11

12 An example of easy to implement set of rules is: Sell Short a strangle on 30Year - US Treasury Bonds using contracts with closest monthly expiry date, with both legs at Delta = 0.1, and let the trade survive until maturity. Then open the next trade using the same rules. Rules like these are easy to implement in the Option DBE, thanks to the possibility of accessing historical option chains that make strike selection very easy. We are currently working on the implementation of our own programming language. This language is being made in order to look like TradeStation s EasyLanguage, but with all of the features needed when dealing with options. When it will become available, backtesting of strategies like the one just mentioned will be operated in the matter of minutes. 12

13 4. System Implementation To show how to implement a system we will take as an example the methodology stated before: Sell Short a strangle on 30Year - US Treasury Bonds using contracts with closest monthly expiry date, with both legs at Delta = 0.1, and let the trade survive until maturity. Then open the next trade using the same rules. This will be our system of choice for every future examples. Cadence, Symbol and Dates The first thing to do is to select the Cadence of the options and the Symbol of the underlying futures needed. In our case Cadence is Monthly, and underlying symbol is US. Next step is to define the starting and ending dates of the test. Let s assume we want to start our test on May 2012 and run it until May Once the date of the first trade is set, the list of available Expiry Dates is updated. From now on, every time the Quote Date is changed the Expiry Dates list will be updated. We want to let our trade run until maturity and then we want to start the following trade on the same day. Thus, it has been natural to select as starting date for the test, the maturity date of the option expiring on May

14 This is why the first element of the Expiry Dates list is labelled with Maturity, to highlight the fact that the Quote Date selected is also the Maturity Date of some contracts. Since we want to sell short a strangle on the first available maturity we select the second item of the list, as just said we ignore the first item because there s no life remaining for those contracts expiring on that same day. Strike Selection Once an Expiry Date has been selected two things happen: first the list of available Strike Prices is updated, second the historical Option Chain is elaborated for the contract expiring on the selected date. In order to create our trades we now have to select the strike price needed. There are two ways of achieving this: 1) By selecting the strike price from the list using the main control panel. In the list, available strike prices are sorted in ascending order, with Call Moneyness shown between brackets. Call Moneyness is useful to get a quick idea of how far from the At The Money contract the strike we are reading is. 14

15 After the strike is chosen the main control panel becomes available. This happens because we gave to the program all the information it needs to identify an option trade: an opening date, the expiry date of the contract we want to trade, a strike price, and the option right. By default, the selected option right is Call, to change it click on Put in the Right block on the right of the strike selection drop down menu. Now, by clicking on Buy or Sell in the lower right corner, we can create an historical trade. 2) The second way of selecting the strike price consists in choosing the contract we need to trade from the option chain. This option is often better because the option chain carries a lot of information that is usually taken into account when trading options. To open the option chain, click on View Chain button on the right-hand side close to the expiry date selection drop down menu. When the chain is loaded, this is what will appear: We will cover the details of the Option Chain window in a later section. For now we will just use it as a tool to select and open historical trades. By default the information shown include: Price, Delta, Implied Volatility, Volume and Open Interest for both Call and Put contracts. Since our strategy was to sell a delta 10 strangle, what we should do is to find the call and put strikes with delta closest to

16 After we found the correct strike all we have to do is right click on any cell of the line and select the Sell Option: The $154 strike price contract carries a delta of 0.11, and therefor is the closest strike to our needs. The price of the option is shown in USD unit term. To see the actual whole value of the contract, which in the US case is unit price times 1000, flag the Dollar Value label in the upper left corner. Backtest Box After having sold both legs of the strangle, in the backtest box, two new trades will have been created as shown in the figure: The size of the trade can be changed by double clicking on the cell in the column Contracts, and manually changing the number of contracts traded. Now, before the outcome of the trade can be calculated an Exit Logic has to be chosen. 16

17 5. Exit Logics For now, there are 4 different ways an option trade can be closed: Time Exits 1) Wait until Maturity. The trade will run until Maturity, then autonomous exercise is assumed to occur. 2) Exit on a specific date. The trade will be closed on a specific date between the opening date and the maturity date. These first two represent a time limit after which the position is closed. Once a time exit is set, an intermediate exit can be setup to close the position before the time exit occurs as shown in the next paragraph. Profit Target and Stop Loss 3) Stop Loss. The trade is closed when a stop loss level is reached, based on one of the logics available. 4) Profit Target. The trade is closed when a take profit level is reached, based on one of the logics available. To setup a time exit for the trade, click on the drop-down cell in the column with header Time Close. The list will show all available days between opening and maturity. Click Maturity to let the trade run until expiration date. As soon as an exit date is chosen the trade is calculated and profit registered. At this point a Stop Loss and a Profit Target can be set up. Currently there are 3 different ways available to define exit levels. 17

18 These 3 ways are analogous for Profit Targets (TP) and for Stop Losses (SL), in the first case they will identify a level of profit at which the trade will be closed, in the latter they will identify a level of loss instead. 1) % of Open Price. The trade is closed when the option premium moves of a certain % in profit (TP) or in loss (SL). For example, our first trade was a sell short call, and we sold it at a premium of $ ( $ Dollar Value). If we set up a SL of 80%, since we are selling short, our SL will be set at an 80% appreciation: x ( ) = $ ( $ Dollar Value). If we set an 80% TP, since we are selling short, the trade will be closed when premium reaches an 80% depreciation: * (1 0.8) = (53.125$ Dollar Value). 2) Price Move. In this case, the trade is closed when the option premium moves of a certain amount expressed in price points. Let s take the same example as before, selling short at : a Stop Loss of 0.2$ means that the trade will be closed after a 0.2$ appreciation: = $ ( $ Dollar Value). A 0.2$ Take Profit would result in the trade being closed after a 0.2$ depreciation: = $ (65.625$ Dollar Value). 3) $ Amount. A Stop Loss or a Take Profit stated in $ Amount provides a threshold expressed in terms of actual trade profits and losses. A 200$ TP would result in the position being closed after the trade goes positive of at least 200$, while a 200$ SL would close the trade after a 200$ loss. This kind of TP and SL are size sensitive: the same amount of TP and SL, in a trade with double number of contracts traded, would result in half available space, because of profits and losses being built at twice the speed. To setup a TP or a SL two inputs must be provided: Mode and Value. To select the mode extend the drop down menu in the column TP Mode (SL Mode for Stop Loss), in the row of the trade for which a TP or SL must be set. Then enter a value in the cell on the right representing the entity of the TP/SL. After the mode has been selected, and a value entered, the trade is automatically recalculated, profit updated and a string in the Comment column will tell us if the trade was closed because of the SL or the TP, and in case it was, when it was closed. 18

19 6. Automation and Defaults Option Language Since in some case many hundreds of trades will have to be created, the process of manually entering all the historical trades of the backtest can be a long one. We believe that it s a fair cost for the information achieved, no other tool allows the user to get this kind information so easily about Options on Futures trading, but we nonetheless think a higher degree of automation is needed. We are currently working on developing an in-house programming language, based on TradeStation s EasyLanguage, that allows the user to automate the entire testing process. This is what a full test looks like: Entering all those trades is not an easy task, and it could require a couple hours of work. The same test, starting from the next release of the program, will be done a lot quicker. It will be enough to enter this code block in the editor to create the strategy, and then launch it: 19

20 While this feature is still not available, it s currently our top priority, and we are looking forward to releasing it as soon as possible. Defaults Now, by the way, to reduce the workload, the user can set some default parameters. This default parameters are useful when a static logic is used to define the exit criteria of trades. In our example, our basic rule was to exit our trades at Maturity. Defaults allow us to tell the program we are going to use always the same exit logic, and thus not to wait for further input before calculating the trade outcome. To set up a default is very easy, they are located in the lower half of the control panel as shown in the figure: In our case, we could have just set the Time Close default at Maturity: By doing this, after having executed an historical trade by pressing Buy or Sell, the program will automatically calculate the trade s outcome based on a wait until maturity logic. Default TP and SL work just as non-default ones. In the Defaults, as a time exit, we have as well the option to define a number of days prior of expiration as a time limit of the trades. This is done by selecting, under Time Close default, N Days, and by providing the number of days wanted in the textbox on the right. Defaults are also useful for trying different exit logics in almost no-time. 20

21 After all trades have been entered in the backtest box, to change their exit logic is a quick and easy task. Just set a new default and press the button with the two circled arrows on the right. All trades will be recalculated using that default: This feature allows the users, while waiting for full automation to be implemented, to test different settings with ease. Defaults can be set up in the Option Chain window as well, their location is in the top panel as shown in the figure: Once set up, after the window is closed, the defaults will be remembered when the Option Chain is opened the next time, even if on a different contract. 21

22 7. The Option Chain Window This is a full view of the Historical Option Chain: The Option Chain Window is a very useful tool to gather extensive historical information. For every symbol available, every maturity date, every single trading day, an historical option chain can be displayed. In the Windows Bar, symbol, quote date, and expiration date of the contracts are shown. This tool allows the user to choose which information to show and which one to hide, this makes it easier to keep the window clean and to not overload it with data that is not needed. 22

23 The Option Chain, for every contract historically available, can show: - Price - Delta - Gamma - Vega - Theta - Rho - Implied Volatility - Volume - Open Interest By default, when opened, the chain will show Price, Delta, Implied Volatility, Volume and Open Interest. Volume and Open Interest columns are coloured according to a logarithmic-scaled quantile succession, to make information easier to access. Price, Volume and Open Interest cannot be removed from the data table, it is at user s discretion to choose what else to show. Flag what is needed in the Calculate box as shown in the figure, then press Recalculate: After navigating the chain, which can be quite long, to go back to the At The Money strikes press ATM button in the Calculate box. The Dollar Value checkbox, in the upper left corner, allows the user to show option premiums expressed in contract size instead of per unit. Underlying Price field shows the settlement price of the underlying on the date the chain is being calculated. The Show box allows the user to switch between Intrinsic Value, Extrinsic Value and Whole Value of premiums. Defaults have been discussed in a section 6. Group Trades allows the user to group trades opened from the Option Chain, the grouping feature will be discussed in section 11. The Action box allows the user to choose if to highlight a certain level of delta. We think this feature can be useful since delta is often used as a mean to select price levels of strike prices. 23

24 To highlight a certain delta level within the chain, flag Highlight Delta, and input in the Call and Put textbox the levels needed as shown in the figure: Follows a screenshot with corresponding output: After closing and reopening the chain window, the Action box and Defaults box inputs will be remembered. As discussed in section 5 trades can be opened from the chain, by right-clicking on the line of the contract to be traded, and selecting whether to buy or to sell it. 24

25 8. Price Modelling and Database Adjusting An option database, even for a single symbol, is massive. To get an idea of the amount of information, think that the full database is composed by more than half a billion lines. Such a huge size, of course, has made it difficult to keep its quality at 100%. Our database is highly precise and reliable, especially in more recent years (since 2010) where holes affect less than 99% of available data. For older years, instead, it becomes a little bit more frequent (but still extremely unlikely) to find holes. Fortunately, the clear majority of these holes occur at the date of contract expiration. It happens because sometimes the just expired contract (on that same date) was not recognised as a trading contract on that day. This kind of holes are the easiest to be fixed because the option premium at maturity is deterministic, and it equals to intrinsic value. This is the first option for database live fixing. By Flagging Correct Premium at Maturity with Moneyness, in the Defaults box, when the software cannot find a premium at maturity, because it s missing in the database, the Option Database Explorer models it using the payoff at maturity of the contract in object. The payoff of call options at maturity is defined as follow: P = Max 0, S K Where S is the Underlying Price at Maturity and K is the strike price of the option contract traded. K S is moneyness for Puts. In the rare occasions in which a hole is found far from expiration, a more complex solution is used. By flagging Model corrupted data with Black, Sholes equation, in the Defaults box, the user allows the Option DBE to model missing data using the Black and Scholes approach. When modelling option data using the Black and Sholes equation a few assumptions have to be made. The most important is how market expectation of future volatility is defined. This is not a trivial question. Future volatility expectations vary among different strike prices, growing toward the extremes. Another problem is that usually a hole in the database affects all strike prices, on a certain date, for a given maturity date. 25

26 These facts make using the implied volatility of contracts with different strike prices not a viable option. We decided to model today s volatility using the average of yesterday s and tomorrow s implied volatilities of contracts with same moneyness. There is no closed solution of the Black and Sholes equation in the form σ = f C / C / being the known Call Premium and Sigma the Implied volatility, thus a numerical approach is used 1. After an estimate for σ is found, it is used to solve the Black, Sholes equation reduced to its closed-form solution for European Options (to solve for Put Options put-call parity can and must be applied): C σ = N d 2 S N(d 4 )Ke 789 d 2 = 1 σ T ln S K + r + σ4 2 T d 4 = d 2 σ T Where: N = the cumulative distribution function of the standard normal distribution S = Current Underlying Price K = Strike Price T = Time to Maturity r = Annual Risk Free Rate of return (continuous compounding) σ = Current Market expectation of future volatility To obtain maximum result reliability, we suggest keeping both options flagged at all time. 1 Newton-Raphson method is used to find successively better approximations to the roots of the equation σ = f(c / ). The method implementation is as follow: σ BC2 = σ B C σ B C / E FG Where σ B is the i-th estimate for implied volatility, C σ B is Black, Sholes option price calculated in σ B ; C / is the know option price; I J I KG is the partial derivative of the Black, Sholes premium with respect to the current estimate of implied volatility. The i-th estimate is assumed to be Current Implied Volatility when: C σ B C / C / Which implies a precision of 0.1%. Initial guess for σ B is determined through Bharadia, Christofides, Salkin estimate (Bharadia, Christofides, Salkin, 1996): σ N = 2π T C / p S p Where T is time to maturity, S is current underlying price, and p = Q7R 4. 26

27 9. The Performance Report Once all trades of a test have been executed, the Performance Report can be opened to gather information about the results of the test. Here, both performance statistics and charts can be found. To open the Performance Report click on Test -> Open Performance Report: This is how the Performance Report looks like when opened: The chart shown is the Close-To-Close, trade-based, equity line of our strategy. The lower chart shows profits and losses of each trade. In this case, since we are selling naked strangles, profits are very frequent and closely distributed around the mean, while losses are less frequent, sparser, and of higher magnitude. To switch to a daily chart, click on Options -> Chart Type -> Equity Detailed. 27

28 The chart will now show a date-based charts: The lower charts, this time, shows profitable and non-profitable days. The lower chart can be used to show profit/losses of the upper chart, or DrawDown. To switch to DrawDown click on Option -> Lower Chart -> DrawDown: As a result DrawDown will be shown: 28

29 The second tab, Performance, gives key statistics about the backtest: The third tab, Test Data, shows the Test Name, the Defaults that has ben applied, and symbols used. This helps to contestualize the performance report when more than one will be opened at the same time. 29

30 10. Grouping Trades When trading options, often, more than one of them are traded simultaneously in some combined way. This was the case, for instance, with our short strangle system. Strategies can be of any degree of complexity and comprise an undefined number of contracts at the same time. Nonetheless, these multiple trades are often considered by traders as one. To tackle this issue, we introduced so-called Trading Groups. Trading groups allows the user to tell the program, when calculating performances, to treat multiple trades as a single one. Grouping trades with the Option DBE is extremely easy. Flag the Grouping checkbox in Group Trades section as shown in the figure: Select a group number. All trades opened afterwards will be grouped together, to start building a new group it s enough to change the group number. When calculating system performance all grouped trades will be considered as one. It s possible to group trades opened through the Option Chain as well, following the same instructions as above, in the Group Trades section. 30

31 11. TS Code Generator for Naked Shorts Hedging The TS Code Generator is a feature that allows the user to automatically create TradeStation strategies that hedge Naked Short Sell option trades. Once implemented these strategies have to be launched in TradeStation on the appropriate underlying instrument. The outcome can then be visualized in two ways: it can be imported in the Option Database Explorer, that will elaborate it in conjunction with the option strategy as a portfolio (Option + Underlying), or it can be evaluated directly within TradeStation with the performance report. Another way would be to export both the TradeStation Report of the hedge and the Performance Report of the Option DBE into an Excel spreadsheet, and then to aggregate them using a macro. To generate the TradeStation EasyLanguage code of the naked shorts covering strategy click on Edit -> Export EL Code Future Hedging. Once clicked on Export EL Code Future Hedging an option window will open asking for a few information before starting the process. Here the user can choose the name to give to the.txt file where the strategy will be written, on which symbol of those used in the backtest to create the hedging system, if only one symbol was used, as in most cases, only one option will be available. You can ask the program to automatically open the.txt file straight after code generation by flagging Open after Export. The Code Generator will produce a strategy able to cover all and only the option short trades on the selected symbol. Click OK and the Option DBE will start building the code, which will be exported in a.txt file in the installation folder at the directory Exports/EL Auto-Generated Code. To easily access the installation folder, click on File -> Open Data Folder. 31

32 If you flagged the Open After Export option the.txt file will automatically open, showing the code just generated. To use the strategy just generated copy the whole text contained in the.txt file (ctrl+a, ctrl+c), and paste it within a strategy appositely created in TradeStation. Hedging Strategy Code Let s see now the code sections of the hedging strategy. 1. Input and Control Section The first section, until line 33, contains variables declaration. These variables can be used to customize the covering strategy according to the user s needs. 32

33 At Line 8 a comment reporting the location of the.txt file containing the generated code can be found. At line 13 the user can choose a monetary Stop Loss to be imposed to every cover trades. It will be individually applied to all open trades, not to the whole position, when multiple trades are floating at the same time. The inputs related to the long part of the cover (covering short calls) are located at line 16, let s discuss them one by one: - Delay: the distance from the strike price at which the cover has to be placed. It can assume values >, <, and = to 0. In case of a delay equal to 0, the cover trade will be placed exactly at the strike price, in case of a negative delay the cover order will be placed in the Out of The Money area, in case of a positive delay it will be placed in the In the Money area. Thus, for example, if I have a naked short call position with strike 10$, and set a delay of 1 (also expressed in $), the Stop Buy order will be placed at 11$. - SizeMultiplier: this variable allows the user to cover only a fraction (if lower than 1) or multiple times (if higher than 1) the option position. In fact, the strategy is coded in a way that makes the cover trades be opened with the same size of the option trade to be covered. If we want to give double exposure to our cover trade, we will have to set the size multiplier at 2. If we only want a half coverage we will have to input 0.5, by default this variable is set to 1, for a neutral (at maturity) kind of hedge. - NConIfMultZero: if instead we want to disentangle the size of the option trades from the size of the cover trades we will have to set the SizeMultiplier = 0, so that the proportional component of trade size calculation is 0, and NConIfMultZero = X, where X is the number of contract that we want our cover trades to be opened with. If SizeMultiplier is not equal to 0, any value assigned to NConIfMultZero will be ignored. (The name of the variable comes from Number of Contracts to be used If Multiplier is Zero). To better understand SizeMultiplier and NConIfMultZero variables, think of them as related to the cover size in this way: S T = N + M S U Where: S T = Size of the Cover Trade on the Underlying S U = Size of the Option Trade to be covered M = SizeMultiplier, managing proportional section of size equation N = NConIfMultZero, managing constant section of size equation The following constraint applies, independently by user input: M 0 N = 0 33

34 - AllowSideBuy: this variable allows the user to inhibit all cover trades for the call side. The long part of the covering strategy will be deactivated. We ll later see that it is possible to create custom filters, that if set to False would have the same effect. The difference is that filters eliminate the chart lines of the deactivated trades to give the user a feedback about filters operations, while using AllowSideBuy to deactivate the whole long side will result in the chart line to still be placed on the strike levels. At line 17 the filter variables are declared and initialised to true. We do not advise to interact with this variables at declaration stage, filters logic will have to be implemented in a later section, at line 24. At line 24, in fact, we find the same 3 variables again being set to True. By substituting True with any expression after the = sign, it is possible to build a filter that will operate on the side related to that variable. For instance, if we write at line 26: FilterBuy1 = Close > Close[1]; The long cover will only function if current underlying price is higher than close price of one bar ago. It is possible to set three filters at the same time, if we want to leave one of the three filters inactive it will be enough to leave it set to True as by default. Filter 1, 2, and 3 are applied with an AND logic, thus they will have to be all True at the same time for the cover to function. Any OR logic will have to be implemented within the same variable, for example: FilterBuy1 = (C > C[1]) OR (H > H[1]); At line 20 and 21 are located the equivalent control variable for the Sell side, at line 31 to 33 are set the custom logics of the sell side filters. 2. Internal Prelim Calculation 34

35 Here, internal calculation needed by the program to work correctly are performed. For a proper functioning of the strategy we advise not to modify the code contained in this section, from line 36 to Trading Core The actual trading core of the strategy, where entry and exit conditions for covering trades are coded, is here. We advise not to change this section, because it would be difficult to modify it consistently in its entirety. This section, in fact, consists of 11 code lines for every option that has been short sold, it would be impractical to operate a change that stay consistent on every trade. 4. Stop Loss and Performance Export In this example, this block starts at line 1314, a number that varies depending on the number of option trades that have to be covered. The stop loss is set up in line 1315 and In the block from line 1319 to line 1328 the result exporter is coded. 35

36 The folder selected as destination for the export of the results of the hedge is automatically set at the directory Exports/TradeStation Equity Export within the installation folder of the Option DBE, to change it the user has to modify line 1323 accordingly. The name of the.txt file that will contain the result is chosen at line 9, over the input block, at variable TestName (if modified make sure the name ends with.txt). To enable the export the block has to be uncommented, to do so the brackets at line 1320 and 1328 have to be deleted. The result is exported with a format that makes it comparable with the option backtest, it consists of a daily equity. Launching the strategy The figure above shows what will appear on the chart after having compiled, loaded and run the strategy on the appropriate underlying instrument. Coloured lines will be automatically generated. Solid green lines show the strikes of the call options sold (meaning stop buy order will be used as covers), while solid red lines show the strike prices of put options (meaning stop sell order will be used as covers). Dashed lines show the actual level where the stop orders are placed, 36

37 this level is not equal to the strike price when Delay input is not set to 0. In this example, Delay was set to 0.5. If more than one option has to be covered at the same time strategy properties (at Properties for All window) have to be changed accordingly. More specifically Position Limits have to be set as in the figure below: This is how a two puts trade would be covered: As shown by the solid red line, there are two puts sold, with different strike price, at the same time. They are covered by different trades, that are managed separately as can be seen by the fact that the stop loss exit of the first cover trade of the lower put does not affect the previously opened trade. 37

38 Confronting Results Aggregating results and analysing performance of the combined portfolio of the Option Strategy and the hedging strategy is easy. First of all, the hedging strategy must have been executed on the underlying, in TradeStation, with the block containing the exportation code uncommented. This operation will generate a result summary file, in.txt, saved in the installation folder of the Option Database Explorer. Then, in the Option DBE, in order to aggregate results, the.txt file has to be imported. To do this click on Edit -> Import TS Report, as shown in the figure: A dialog, opening automatically at the default location (Exports/TradeStation Equity Export), will let you open the file. Here the user will have to choose the report he wants to import. The importing process should be fast, in the Log section the label TSReport Imported tells the user that the operation was completed successfully. Done. Now the hedging equity line is in memory, and the Performance Report, in Detailed mode, will look as follow: 38

39 The three lines represent the equity line resulted from the Hedging strategy (red), the equity line of the Option strategy (blue), and the portfolio equity line (black). The Close-To-Close equity line will still only show option trades because CtC equity lines are not aggregable with sufficient precision. In Performance tab are now available key statistics concerning the option strategy, the hedging strategy, and the combined portfolio. Usual CtC performance metrics are also shown. In order to eliminate the last imported report from memory click on Edit -> Reset TS Report. 39

40 12. Exporting Results Backtests produced with the Option Database Explorer can be exported to an excel spreadsheet for further consultation and to operate custom calculations. To export a backtest click on Edit -> Export to Csv By default, the location for exported backtest is within the installation folder at Exports/Csv Exports, but a custom directory can be chosen. After selecting the directory for the export press Save. A.csv file will be generated, it contains information about trades, cumulative CtC equity line, and the detailed equity line. 40

41 13. Miscellaneous Test Name Test Names are used to keep track of backtests in a friendly way. They can be any sequence of characters, and, to be set, a Test Name has to be input in the Test Name bar, as in the figure: Once entered it will be associated with the test, and saved with it. This means that it will be remembered once the test is closed and reloaded. It will always be shown in the Test Name bar, so that the user can easily remember what backtest he is working on, and it will also be shown in the Performance Report, in the Test Data tab. If a custom name is not entered, by default, the Test Name will be the name given to the save file of the backtest. Save and Open a Backtest It is possible, when needed, to save the work done. After trades have been executed, go to File -> Save As. Default directory for save files is in the installation folder at /Save Files, but a custom location can also be chosen. Once saved, all the information contained in the backtest box will be saved, along with Test Name and Defaults used. To open an old backtest click on File -> Open. Clear All If needed, in order to cancel all trades in the backtest box and clear the working space go to Edit -> Clear All. Warning: all information will be lost, if then the test is saved and file overwritten there won t be any way to recover it. Data Folder All information is stored in the installation folder: the database is stored there, and it s the default location for save files and exports. It happens quite often to need to go to the installation folder. To make it quicker, we put a button in the File menu, called Open Data Folder, that allows the user to quickly open a File Exploration window directly in the installation folder. 41