Strategies for the Simple Timing Indicator (STI) and MACD to Improve Investment Performance AFTA October 18 & 22, 2011 Jim Gunn jimgunn@jgunnresearch.com
Disclaimer No Representations are made that the investment strategies presented herein or discussed by the presenter are complete, without error, or will produce desired investment results. This presentation is made for information sharing purposes only. Nor are any representations made that the investment strategies discussed have been employed to achieve successful investment results. Many of the results are based on back testing with historical data and simulated investing. Future investing results may significantly differ. You should perform your own independent research and “Due Diligence” prior to making any investment decisions, and invest accordingly. 10/22/2011 2
AAII Technical Indicator Quote “AAII Computerized Investing”, 4 th Qtr 2011, p1, “Building AAII Stock Screens” Article, by Wayne Thorp “All of the stock screens AAII develops are based on our interpretation of an approach, often as it was outlined in a book, article or academic paper. For example, in this issue I finish my two-part discussion of Charles Carlson’s formulas for selecting promising dividend-paying stocks. In developing this and other stock screens, I follow the author’s approach as it is outlined in the source. However, many successful investors do not want to “give away the farm” when writing about their approach, so they are purposefully vague when outlining their search criteria. Therefore, there are times when we must use our investment expertise, and the wisdom of other market gurus, when deciphering what the author is trying to say.” Lessons Learned • Investments Strategies offered by “Experts” and others are often incomplete; sometimes purposefully or sometimes due to changing market conditions • No Technical Indicator or combinations of Indicators or strategies are reliable over all time frames, selected investments, and market conditions • Best investment successes will usually be achieved by developing and adapting our own information, strategies, due diligence, study, and self-training 10/22/2011 3
Presentation Topics • Overview of Excel as Investment Tool • Overview of use of STI and MACD Strategies Explanations • Simple Timing Indicator (STI) • Moving Average Convergence Divergence Indicator (MACD) • Where indicators usually work and don’t • Human Chart Pattern Recognition • Weekly Chart Examples • Seasonality Analysis • Interactive Charts of Results using Excel (Tuesday, as time permits) • QQQ – Powershares NASDAQ 100 Index ETF • SPY – S&P 100 Index ETF • IJH – iShares S&P 400 Mid Cap Index ETF • IJR – iShares S&P 600 Small Cap Index ETF • AMZN – Amazon • EMA vs SMA Analysis • Above and Audience Requests (Saturday at Baylor) • Advanced Inputs Appreciated to load data base, if necessary (and possible) • Limited others possible, but more time consuming to download data • ~1200 Symbols in current data base 10/22/2011 4
References • Appel, Gerald; Technical Analysis; MACD Chapter 8; 2005, FT Press • Elder, Dr. Alexander; Come Into my Trading Room; 2002, Wiley • Elder, Dr. Alexander, Presentation at Money Show Dallas, June 2011 • Richards, Fred, STI discussions in various Presentations at Dallas Investors Forum, 2009-2010 • Saddler, Rick; “Swing Trade the T-Line” Presentation, DFW AFTA, October 10, 2010 (T-Line is the 8 EMA of Close) 10/22/2011 5
Methodology and Tools • Goals: • Better understand Indicators and their utilization in investing • Utilize Black Box Indicators, but better understand when not reliable, and adapt • Study Historical Charts of Investments of Interest • Carefully note when patterns are most reliable and least reliable and adapt • Judiciously apply human chart pattern recognition abilities and when appropriate follow most reliable indicator(s) or combinations • Use Excel for all charting, indicator calculations, financial analysis, back testing, step forward analysis, seasonality analysis, etc. • Excel/Office 2010, released in June 2010 • Can use other tools, such as stockcharts.com, yahoo finance, TC2000 etc. for daily strategy(s) execution, but these often have deficiencies for back testing, historical data (especially longer time frames), step forward analysis, etc. • Many Stock Indicator “How to” papers often give brief discussions • Excel examples often provided to demonstrate calculations • Many indicators are old and based on pre-computer era computing and plotting resources • Simplifications were often used to reduce calculations or plotting efforts • Modern Computer and Communication technologies offer potential to enhance use of indicators 10/22/2011 6
PS QQQ (NASDAQ ETF) (last 2+ months) 8EMA 20EMA 50EMA 200EMA PowerShares QQQ Trust, Series 1 Price 60 20 EMA 8 EMA 50EMA 200EMA 58 56 54 52 50 STI: Long when 8EMA >= 20EMA or equivalent when MACD >= 0 48 7/31/11 8/7/11 8/14/11 8/21/11 8/28/11 9/4/11 9/11/11 9/18/11 9/25/11 10/2/11 10/9/11 10/16/11 10/23/11 Vol Vol SMA QQQ Volume 220 200 180 160 140 120 100 80 60 40 7/31/11 8/7/11 8/14/11 8/21/11 8/28/11 9/4/11 9/11/11 9/18/11 9/25/11 10/2/11 10/9/11 10/16/11 10/23/11 QQQ MACD (8,20,7) Histogram DIF Sig 1.5 1.0 MACD(8,20) = DIF = 8EMA – 20EMA 0.5 0.0 -0.5 -1.0 Histogram = DIF ‐ SIG -1.5 -2.0 7/31/11 8/7/11 8/14/11 8/21/11 8/28/11 9/4/11 9/11/11 9/18/11 9/25/11 10/2/11 10/9/11 10/16/11 10/23/11 10/22/2011 7
Investing Rules • Primary Indicators: • MACD(8,20,7) of Close • Long MACD > SIG; • Short or Cash MACD < SIG • Simple Timing Indicator • Long 8EMA >= 20EMA or MACD >= 0 • Short or Cash 8EMA < 20EMA or MACD < 0 • Secondary Indicators • T-Line: (Especially note support or resistance over time) • Long: C i >= 8EMA (Support) • Short or Cash C i < 8EMA (Resistance) • Supporting/Confirming Indicators (Maintain Awareness) • EMA trending Up (Bullish) or Down (Bearest) • MACD trending up (Bullish) or Down (Bearest) • MACD Histogram trending up (Bullish) or Down (Bearest) • Money Management • Limit loses: 6% - 8% 10/22/2011 8
Excel Block Diagram: Data, Storage, and Calculation Excel/2010/VBA Charting Indicator Calculations Control Panel(s) Active Symbol data, MA, MACD, & Financial Analysis VBA Statement Seasonality Analysis to download QQQ Workbooks.Open ADO ("http://ichart.finance.yahoo.com/ table.csv?s=QQQ") to Access Data Yahoo Web Access/2010 Stock Prices & Over 1000 Symbols: Volume Price: Open, High, Low, download Close, Volume Data Storage 10/22/2011 9
Functionality Additions since AFTA Dec 2010 Presentations 1. Multipage Menus 2. Menu Setting STI & MACD parameters (e.g. 8 EMA or 10 SMA, etc.) 3. Support for Stock Splits 4. Enable Re-charting with select data update a. Sub-Time Frame Results in back testing b. Stock Splits 5. Weekly Charts 6. Menu Driven Daily Price Updates from Yahoo 7. Increased Symbols in Data Base a. S&P 500 b. NASDAQ (Market Capitalization > $900,000 c. Working Database for Indexes, ETFs, Mutual Funds, others, etc. 8. Step Forward Analysis 9. Seasonality Analysis 10/22/2011 10
Stock Price/Volume Data Yahoo Data Source • www.finance.yahoo.com • End of Day (EOD) Data • EOD data typically fully available by ~9 – 10 pm on trading days • Data Downloaded: Date, open, High, Low, Close, Volume, and Adjusted Close • Only Adjusted Close accounts for splits and dividends • Initial download of all historical data for a Symbol requires a special setup in automation • Subsequent daily updates via automation • Mutual Funds, indices, sectors, etc. • ~1200 Symbols in current data base • 8000+ Symbols can be accommodated in Access, but download time would be long 10/22/2011 11
Back Test Methodology • Use End of Day (EOD, Close) Data • Make next day’s Long, Short, or Cash Decisions with EOD indicators • Go Long, Cash, or Short with Open of next Trading Day • Note: • Can be more flexible with human intervention • all discussions today will use next day open for buy/sell 10/22/2011 12
Exponential Moving Average of Close at time i �EMA i � EMA � � EMA ��� � � � � C � � EMA ��� � of Close where C � is the close on the latest date; C ��� latest Trading Day minus 1 For EMA of other indicators, e.g. Volume, Use V i , instead of C i � should be less than 1 and greater than 0 � � � = ���� (typical) or � = ���� where 20 is the length of a 20EMA � Note: 20 SMA i = ∑ � � ���� Observations • EMAs typically take 2 – 3 times the period to stabilize • For example, a 20EMA can require 40 to 60 historical calculations to converge • Simple Moving Averages (SMA) may be employed, but EMAs’ respond faster and usually provide better performance. • MA indicators generally are more reliable that MACD indicators, but catch market changes slower that MACDs. • MA have whipsaws, but generally less frequent than MACD 10/22/2011 13
Recommend
More recommend
