Скачать презентацию Delving into Vector Vest 7 2012 Jim Tebay Скачать презентацию Delving into Vector Vest 7 2012 Jim Tebay

56cd75b4355b3e4b3a214122fe811c78.ppt

  • Количество слайдов: 82

Delving into Vector. Vest 7 2012 Jim Tebay jimt 914@comcast. net Delving into Vector. Vest 7 2012 Jim Tebay jimt [email protected] net

Disclaimer • Although every precaution has been taken in the preparation of this presentation, Disclaimer • Although every precaution has been taken in the preparation of this presentation, the author assumes no liability for errors or omissions • It is offered without a warranty of any kind, either expressed or implied • All concepts should be taken as points of departure for serious individual research into using Vector. Vest 7 -- AND NOTHING MORE • Use of this material constitutes your acceptance of these terms. 2

Trading Profile • My education was in engineering/management • My career was in high Trading Profile • My education was in engineering/management • My career was in high tech. I am retired • As a part-time trader for over 15 years, my stock analysis program was initially AIQ, then TC 2000, and is now Vector. Vest • I have NO affiliation with Vector. Vest, except for using their products • My trading interests are in developing and teaching winning systems. 3

This Presentation’s Objectives • To illustrate means for: – Making money in differing market This Presentation’s Objectives • To illustrate means for: – Making money in differing market conditions – Developing trading plans that work – Reinforcing day-to-day discipline – Improving stock selection through cherry-picking – Using plans and checklists for consistent gains • To reveal lesser known features of VV 7 • To have you leave with useful ‘take-aways. ’ 4

‘You have to learn the rules of the game… and play them better than ‘You have to learn the rules of the game… and play them better than anyone else. ’ Albert Einstein This presentation is mostly rule-based, i. e. , mechanical. Exception examples are cherry -picking or partial profittaking. A symbol (d) appears in slides, which include discretionary material 5

A Favorite Vector. Vest Saying • ‘The secret to making money is knowing When A Favorite Vector. Vest Saying • ‘The secret to making money is knowing When to buy? • A Market (and optionally, a Stock) Timing Question What to buy? • A Search Question When to sell? ’ • A Stop, Market Timing, or Partial-Sell-Off (d) Question • Answers to these three questions create the framework of a trading plan. 6

Making the Questions Meaningful 1. When to buy? • At date of (or after) Making the Questions Meaningful 1. When to buy? • At date of (or after) entry signal from one of the four built-in Vector. Vest’s market timers or from a custom timer (when timing for an entire portfolio) • And, if used, date of an entry signal from a stock timer (when timing for an individual position in portfolio) • And, if used, a follow-through from prior day • Specifically, 10 to 10: 30 AM Eastern, if possible • Or, if back testing, at the outset of a specific lookback period. 7

Questions (Cont) 2. What to buy? • Selected stocks from a built-in or modified Questions (Cont) 2. What to buy? • Selected stocks from a built-in or modified search, or • Selected stocks from Derby Winner searches • Then, from the above, cherry-picked stocks (d) (see later slides re cherry-picking) 3. When to sell? • At date of exit signal from stop loss and/or target stop (for individual stocks in portfolio) • Or, a date from standard or custom market timer (for entire portfolio) • Or, let the portfolio extinguish (as individual positions get stopped out) • Or, through partial profit-taking (d). 8

Another Favorite Saying • ‘Buy safe, undervalued stocks that are rising in price in Another Favorite Saying • ‘Buy safe, undervalued stocks that are rising in price in a rising industry in a rising market. ’ Saying is valid for an up-market • A corollary is also factual: ‘Sell unsafe, overvalued stocks that are falling in price in a falling industry in a falling market. ’ This statement is valid for a down-market • These adjectives in the narration, i. e. , safe, can be reduced to parameters in a search. 9

Modifying Searches • The Vector. Vest 7 contains more than 200 builtin searches. Any Modifying Searches • The Vector. Vest 7 contains more than 200 builtin searches. Any of these searches can be modified by changing or adding parameters and/or by changing the sort • For example, adding a Relative Safety parameter may help to reduce the Maximum Draw Down (MDD) and help to increase the % winners (in a bullish strategy) • An illustrative parameter: RS | => | 1. 0 10

Buy stocks that are ‘Rising in Price’ • A rising stock price can be Buy stocks that are ‘Rising in Price’ • A rising stock price can be uncovered by one of several crossover search parameters – Example: * Stock Price (Split. Adjusted) * Requires Protrader EMA Crossover EMA(12) crossing above EMA(26) within the last 3 days 11

Buy Stocks that are from ‘Rising Industries’ • A rising industry price can be Buy Stocks that are from ‘Rising Industries’ • A rising industry price can be ‘timed’ by using a trending search parameter: Industry Price | Trending | Daily[---++] • In this example, the industry price started trending up two days ago. 12

Buy ‘Undervalued’ Stocks – To discover ‘undervalued’ stocks, add a search parameter such as Buy ‘Undervalued’ Stocks – To discover ‘undervalued’ stocks, add a search parameter such as ‘Value > Price’ – Some traders* may be more restrictive by adding, say, ‘Value > 1. 5*Price’ – Beware! This parameter doesn't improve Bottom Fishing searches, as value is already greater than price in all probability. * Value is a estimated number, which may be somewhat 13 incorrect. So traders may make allowances.

Buy Stocks ‘During a Rising Market’ • A rising overall market is normally signaled Buy Stocks ‘During a Rising Market’ • A rising overall market is normally signaled by one of four built-in timers: Confirmed Call, DEW, Primary Wave, or Green Light Buyer. • A custom timer using, say, a 20/40 MA crossover signal of the VVC may also be used • Sometimes a trader uses only a rising stock price as a signal. He may ignore overall market signals altogether (not recommended for beginners). 14

Support/Resistance • Support/Resistance (S/R) concept asserts that price movement will tend to stop and Support/Resistance • Support/Resistance (S/R) concept asserts that price movement will tend to stop and reverse at certain predetermined price levels • S/R lines are useful in finding the most promising stocks via cherry-picking, using a stock graph (see later slides) • Successful traders avoid upward trending stocks approaching (from below) a recent resistance line or approaching a 52 week high, or vice versa in shorting situations. 15

Campaigns vs. Forays • Vector. Vest uses the word ‘Campaign’ to mean a round-trip Campaigns vs. Forays • Vector. Vest uses the word ‘Campaign’ to mean a round-trip trading venture into the market and later out again. For example, trading throughout a period from a confirmed up signal to a confirmed down signal constitutes one campaign • Some users believe that the word ‘Campaign’ suggests a longer term activity than is present in most trends. They prefer the word ‘Foray’ • In this presentation, the Vector. Vest word ‘Campaign’ is used. 16

Figures of Merit • • Average Annual Rate of Return (AARR) Compound Annual Growth Figures of Merit • • Average Annual Rate of Return (AARR) Compound Annual Growth Rate (CAGR) S&P 500 comparison (a benchmark) Maximum Drawdown (MDD) try to hold to < 20% • No. of trades and % winners • No. of campaigns and % campaign winners • Calmar Ratio compares the CAGR to MDD. 17

S&P 500 Monthly Chart for 10 Years S&P 500 lost 22% over these 10 S&P 500 Monthly Chart for 10 Years S&P 500 lost 22% over these 10 years? % in these 10 years 18

Benchmarks • Wall Street pundits commonly use the S&P 500 as a benchmark for Benchmarks • Wall Street pundits commonly use the S&P 500 as a benchmark for measuring general market action and performance • Vector. Vest uses its benchmark called VVC, a composite average of around 8000 stocks • During the same 10 years (2000 -2010), VVC gained about 6% vs. a loss of 22% by S&P 500. • So, not all market benchmarks are the same or even similar (see next slide). 19

VVC Weekly Price Curve for 10 Years Four Major Trading Opportunities Leg 2 Leg VVC Weekly Price Curve for 10 Years Four Major Trading Opportunities Leg 2 Leg 1 Leg 3 In this instance, these four trends occur over about 70% of the 10 year period Leg 4 20

Making Money in Differing Market Conditions For Educational Purposes Only 21 Making Money in Differing Market Conditions For Educational Purposes Only 21

Educational Studies • The following 4 studies are all conducted automatically using the Vector. Educational Studies • The following 4 studies are all conducted automatically using the Vector. Vest 7 back tester with Autotester and Protrader • Each leg is displayed on the preceding slide 20. No discretionary actions such as cherry-picking or other important, though manual, trading steps can be done in the automatic mode • Any VV 7 user can replicate these studies • The testing period dates are chosen to demonstrate high performances, that is, they are 22 ‘showcased. ’

Bottom Fishing – Leg 4 ‘The Art of Buying Low and Selling High’ Seeking Bottom Fishing – Leg 4 ‘The Art of Buying Low and Selling High’ Seeking to Find Explosive Growth Quality Stocks Beaten Down Stocks Approach Coming off Bottom Use S&P 500 list Use Ascending RT 23

BF (Cont) • When to Buy? Test Period: 3/09/09 to 1/4/2010 Market Timer -- BF (Cont) • When to Buy? Test Period: 3/09/09 to 1/4/2010 Market Timer -- DEW up Stock/Industry Timer -- none • What to Buy? S&P 500/RT Search – Features: Uses good fundamentals stocks, replaces stopped out positions – Feature: Steps in 2 positions per trading day – Sort: RT ascending – Camps: Up only. No action otherwise • When to Sell? 50%G/10%L fixed stop (positions), or when DEW signals down, discontinue replacing positions. 24

Search Aggressive 1 Search Aggressive 1

Bottom Fishing (Cont) • Results – Explosive! – AARR: 293% – CAGR: 343% – Bottom Fishing (Cont) • Results – Explosive! – AARR: 293% – CAGR: 343% – S&P: 79% – No. of Trades & %Winners: 96 64% – No. of Campaigns & %Winners: 11 89% – Maximum drawdown: 14% – Calmar Ratio: 35 * Commissions included but not slippage 26

Bottom Fishing Equity Curve – Leg 4 Green and Red triangles are DEW up Bottom Fishing Equity Curve – Leg 4 Green and Red triangles are DEW up and down timing signals 27

Stock Timing Enhancement • A stock timing parameter, 1 -day MACD Crossover, added to Stock Timing Enhancement • A stock timing parameter, 1 -day MACD Crossover, added to the search, modifies the outcome. It reduces the maximum drawdown (MDD) and increases the % winners, although also reducing the AARR – AARR: – Trade winners: – No of campaigns – MDD: – Calmar ratio: 127% 68% 11 10% 15 28

Stock Timing Search 29 Stock Timing Search 29

Stock Timing Equity Curve 30 Stock Timing Equity Curve 30

RS Study • As noted in slide 10, adding an RS parameter will generally RS Study • As noted in slide 10, adding an RS parameter will generally reduce the MDD and increase %winners. It will also generally reduce profitability • This study adds RS | >= | 0. 9 parameter Basic Study RS Study AARR %: 313 98 CAGR %: 277 92 MDD %: 14 10 Calmar Ratio: 20 9 %Winners: 62 65 31

Value Study • As noted in slide 10, adding a Value parameter generally reduces Value Study • As noted in slide 10, adding a Value parameter generally reduces MDD and increases %winners • This study adds parameter: Value | >= | Price Basic Study Value Study AARR %: 293 107 CAGR %: 343 116 MDD %: 14 8 Calmar Ratio: 35 14 %Winners: 64 73 32

What’s the Best Timer for BFs? Timer CAGR(%) C Call DEW PW GLB 324 What’s the Best Timer for BFs? Timer CAGR(%) C Call DEW PW GLB 324 343 219 70 MDD(%) 14 14 14 6 Calmar Ratio 24 35 16 13 Either the Confirmed Call or the DEW timer works well When in doubt, run comparative tests in out-of-sample periods. 33

Optimization via Scientific Method • The past several slides (pertaining to Stock Timing, Value, Optimization via Scientific Method • The past several slides (pertaining to Stock Timing, Value, Safety, and Market Timing) illustrate a well-known application of the scientific method • That is, one variable is tested while all others are held constant • Caution! Such optimizing may lead to curve fitting (see later slides). 34

Notable Opportunity • BF strategies are probably the most profitable of all, so a Notable Opportunity • BF strategies are probably the most profitable of all, so a trader should take full advantage of them • Major* BFs occur about 2 times per year, with the trends typically lasting about 4 -6 weeks • The longest BF trend in recent times was 14 weeks • In addition, minor BFs occur fairly frequently. * A major BF starts after the MTI reverses from being near or below 0. 6 and the BSR from being near or below 0. 2. 35

Shorting -- Legs 1 & 3 ‘Profiting during a Down Market’ Seeking to Find Shorting -- Legs 1 & 3 ‘Profiting during a Down Market’ Seeking to Find Poor Stocks Reduced Whipsaw Stability Approach RV*EY Ascending Step in slowly Use high Comfort Index & high Relative Safety 36

Shorting • When to Buy? Test Period: Leg 1: 3/00/02 to 10/7/ 02 Leg Shorting • When to Buy? Test Period: Leg 1: 3/00/02 to 10/7/ 02 Leg 3: 6/11/08 to 3/26/09 Market Timer: Confirmed Call (C/Dn) Stock Timer: MACD Crossover • What to Buy: Gravity II Mod (custom) – Features: – Sort: – Campaigns : MACD crosses down over its signal line in the last day; price greater than value; strong CI & RS; replace stopped out positions; step in 3 new positions/day RV*EY Asc (Note: Weak value and earnings yld) Down only. Go to cash otherwise • When to Sell: 30 G/10 L% (individual stock), or when C/Up, go to cash (entire portfolio). 37

Shorting Search 38 Shorting Search 38

Shorting -- Leg 1 • Results – AARR: – CAGR: – Trade Winners: – Shorting -- Leg 1 • Results – AARR: – CAGR: – Trade Winners: – Maximum drawdown: – Calmar ratio 27% 23% 53% 14% ~2 Commissions included but not corrected for slippage. 39

Shorting -- Leg 3 • Results – AARR: – CAGR: – Trade Winners: – Shorting -- Leg 3 • Results – AARR: – CAGR: – Trade Winners: – Maximum drawdown: – Calmar ratio 19% 50% 12% ~2 Commissions included but not corrected for slippage. 40

Shorting Equity Curve -- Leg 1 Shorting Search uses Red Triangle signals 41 Shorting Equity Curve -- Leg 1 Shorting Search uses Red Triangle signals 41

Why CI Indicator? • def: CI indicates a stock's ability to resist severe and/or Why CI Indicator? • def: CI indicates a stock's ability to resist severe and/or lengthy price declines • A high CI parameter, in a search, can reduce drawdowns • See SOTW of 7/31/2009 for more CI details. 42

Shorting Comment • If trading inside an IRA or 401(k), Contra ETFs will need Shorting Comment • If trading inside an IRA or 401(k), Contra ETFs will need to be employed to trade in downturning markets. * These ETFs have a property of going up in value proportional to the market decrease. * A trader is not allowed to short inside an IRA or 401(k) 43

Long – Leg 2 ‘Trading a Nice Uptrend’ • When? Test Period: 3/21/2003 to Long – Leg 2 ‘Trading a Nice Uptrend’ • When? Test Period: 3/21/2003 to 7/12/2007 Market Timer: Confirmed Call Stock Timer: None • What? VST Newcomer with V>P (custom) – Feature: Parameter Value > Price No Step In – Campaigns: Up only. Go to cash otherwise • When? REC <> Buy (individual stocks) or, when C/Dn (entire portfolio). 44

 Leg 2 (Cont) 45 Leg 2 (Cont) 45

Leg 2 (Cont) Nice lower left to upper right curve. Note the flat spots Leg 2 (Cont) Nice lower left to upper right curve. Note the flat spots when portfolio is in cash. 46

Trending Long (Cont) • Results AARR: CAGR: No. of Trade & %Winners: No. of Trending Long (Cont) • Results AARR: CAGR: No. of Trade & %Winners: No. of Campaigns & %Winners: Maximum drawdown: Calmar Ratio: Duration of trend * Commissions included but not slippage 75%* 39% 257 48% 15 80% 19% 2 over 51 months 47

Trending Long Conclusions • Most any trader should be pleased to have a strategy Trending Long Conclusions • Most any trader should be pleased to have a strategy that produces such a high return, accompanied by tolerable maximum draw down, over such a long trend • Several other VST-based strategies are built-in to VV 7 and may produce similar or possibly better results. 48

Just-for-Fun Study • Here are cumulative ‘Just-for-Fun’ gains over the years 2000 -2010. These Just-for-Fun Study • Here are cumulative ‘Just-for-Fun’ gains over the years 2000 -2010. These gains arise by trading the prior four legs cumulatively. When not trading a strategy, portfolio is in cash Leg Opening Closing 1 Bearish $100, 000 $168, 767 2 Long Trend 168, 767 716, 405 3 Bearish 716, 405 891, 893 4 Bot Fish 891, 893 3, 041, 878 49

Just-for-Fun* Study (Cont) • • • Gain: CAGR: Average** MDD: Calmar Ratio (Ten Years): Just-for-Fun* Study (Cont) • • • Gain: CAGR: Average** MDD: Calmar Ratio (Ten Years): Time in trades: 304% 41% 15% 3 ~70% * Called ‘Just-for-Fun’ because it would be highly unlikely that a trader would use these exact strategies. It is the author’s intent only to present a possible, though highly improbable, scenario. See later slides on subject of curve fitting 50 ** An arithmetic average of the MDDs of the four legs

2011 S&P 500 was essentially flat for the year 2011 P 500 1 2011 S&P 500 was essentially flat for the year 2011 P 500 1

Choppy Market ‘Skiing the Moguls’ Seeking Approach To follow Market and Use two timers Choppy Market ‘Skiing the Moguls’ Seeking Approach To follow Market and Use two timers Position movements for each strategy and two strategies To reduce Whipsaw Step in and let portfolio extinguish 52

Choppy Market Trading • When? Test Period: 1/03/2011 to 1/3/2012 Market Timer: Primary Wave, Choppy Market Trading • When? Test Period: 1/03/2011 to 1/3/2012 Market Timer: Primary Wave, both up and down Stock Timer: MACD Crossover • What? MACD Crossover strategies (modified) – – – Feature: MACD Crossover Long Search(mod) for up Feature: MACD Crossover Short Search(mod) for down Feature: Parameter Value > Price added as the mod (for long) Feature: Step in 2 positions/day Feature: Do not replace stopped-out positions Feature: Switch Strategies as Market Timer signals change • When? 20%G/10%L stops (individual stocks) or, when the Primary Wave changes, cash out entire portfolio and refill. 53

Choppy Search (Up) 1 Choppy Search (Up) 1

Choppy Results (Cont) – AARR: – CAGR: * – Winners: – Max. drawdown: – Choppy Results (Cont) – AARR: – CAGR: * – Winners: – Max. drawdown: – Calmar Ratio: – No. of Campaigns: 40% 52% ~10% WOW! ~4 54** * Commissions included but not slippage ** Many campaigns due to choppiness 55

Choppy Mkt. Equity Curve Choppy Mkt. Equity Curve

Choppy Market Conclusions • Often traders dread choppy markets, as the downside risk may Choppy Market Conclusions • Often traders dread choppy markets, as the downside risk may seem to be greater than the upside potential • This strategy’s performance was exceptional in 2011 • The key is a strategy that pick stocks just starting to rise in price as the overall market is rising (and vice versa when shorting) • See SOTW of 5/6/2011 for alternate way to trade in a choppy market. 57

Variety of Strategies • Market Conditions: – Bottom fishing, Two bearish trends, Bullish trend, Variety of Strategies • Market Conditions: – Bottom fishing, Two bearish trends, Bullish trend, and Choppy • Market Timers (all built-in): – 2 Confirmed Call, 1 DEW, and 1 Primary Wave • Stock Timer – Added as a search parameter to enhance performance • Searches: – 1 standard and 3 custom • Stops: – 20%G/10%L, 30%G/10%L, 50%G/10%L, and REC <> Buy • Durations: – 12, 14, 17, and 51 months. 58

Caution • Most important, these strategies may or may not be transportable to another Caution • Most important, these strategies may or may not be transportable to another time period • They may be ‘curve fitted’. Another time frame may not contain price movement enough like the original to get similar results • Do not assume that any SOTWs or especially any contest winning strategies are ‘transportable’ • The longer the back test period, the less likely there will be significant curve fitting. 59

Caution (Cont) • Out-of-sample and longer-period back testing are methods for determining if a Caution (Cont) • Out-of-sample and longer-period back testing are methods for determining if a strategy is robust • Bottom fishing strategies, like S&P 500/RT, are not curve fitted (except for step-in days) and are robust • The other prior three studies probably are ‘curve fitted’ • For example, the choppy strategy does not perform at all well for the year 2010. 60

What to Do? • Since built-in strategies do not necessarily transport effectively, what does What to Do? • Since built-in strategies do not necessarily transport effectively, what does a trader do? • Choose another prior time period, test over longer periods and compare. If similar results occur, OK • Or, use Derby Winners. The safest searches are those matched as both one-day and 5 -day winners. Consider modifying as shown earlier • If the Derby Winners are evenly mixed (some bullish and some bearish), stay on the sideline until market action clarifies. But, if 8 out of 10 are moving in the same direction, go with it. 61

MTI A Powerful Indicator 62 MTI A Powerful Indicator 62

Practical Use of MTI • Use the Market Timing Indicator (MTI) with MACD and Practical Use of MTI • Use the Market Timing Indicator (MTI) with MACD and Views for decision aids on market conditions, entry and exit dates, and stop adjustments • The MTI tends to have a congested and somewhat rounded tops and sharp V-shaped bottoms • When a rising MTI nears or reaches 1. 60, the market is overbought and is looking for a top. When a falling MTI is nearing 0. 6 and BSR is near 0. 2, it’s oversold and is looking for a bottom • The next slide shows MTI with 1. 6 and 0. 6 marker 63 lines.

MTI = 1. 6 MTI = 0. 6 64 MTI = 1. 6 MTI = 0. 6 64

MTI (Cont) • Generally, as MTI rises, buy stocks and replace ‘stopped-out’ positions • MTI (Cont) • Generally, as MTI rises, buy stocks and replace ‘stopped-out’ positions • As MTI nears 1. 60, tighten stops and do not replace ‘stopped-out’ positions. Also, consider selling covered calls. See SOTW of 2/17/2012 • As MTI, BSR, and RT drift down from their tops, protect profits. That is, sell positions (in part or total). Do not add trades. 65

Yet More MTI (Cont) • Upon a C/Dn timing signal or a preponderance of Yet More MTI (Cont) • Upon a C/Dn timing signal or a preponderance of bearish Derby Winners, use a shorting strategy or stay in cash • When the MACD histogram flattens or turns down, tighten stops. If an MACD bearish crossover occurs, consider taking profits or buying puts • See next slide on diverging MTI vs. MACD histogram. 66

Note the divergence of MTI vs. MACD 67 Note the divergence of MTI vs. MACD 67

MTI (Cont) • Switch from MTI graph to Views each day to get guidance MTI (Cont) • Switch from MTI graph to Views each day to get guidance on entry or exit decisions, as well as to acquire the best searches from Derby Winner Selections. 68

Real World Trading • The results of the prior studies were all obtained by Real World Trading • The results of the prior studies were all obtained by auto back testing. VV 7 has an excellent back tester but it is limited in certain ways • A seasoned trader would also do the following: – Cherry-pick stocks (d) before trading by studying their graphs (see SOTW of 3/30/2012 and appendix for full description on how to cherry-pick). – For determining follow-through, verify that the opening of today’s VVC exceeds yesterday’s close (for up-trend) and vice versa 69

Real World (Cont) – Enter broker limit orders after 10 to 10: 30 AM Real World (Cont) – Enter broker limit orders after 10 to 10: 30 AM if a followthrough. Set stop losses right after the order is filled – Look at portfolio positions each EOD. Adjust stops using stock volatility (d) – Sell (cover) any stocks not creating 1% gains after 3 trading days (d) – As the up-trend momentum decreases (as seen by MACD), cease replacing ‘stopped out’ positions. Tighten stops, profit -take or go to cash before a C/Dn (d) – Use margin during instances of a well-established up-trend (d) 70

Ideal Graph Smooth Rising Price Curve Value > Price Low Daily Range, No Spikes, Ideal Graph Smooth Rising Price Curve Value > Price Low Daily Range, No Spikes, No Gaps Rising EPS Value See appendix for practical Graph Settings for Cherry-Picking 71

More Real World • Apply the 10% catastrophic portfolio and 1% position risk rules* More Real World • Apply the 10% catastrophic portfolio and 1% position risk rules* • To reduce whipsaw losses, step in 2 0 r 3 positions per day into a new portfolio, if using DEW, Primary Wave, or Green Light Buyer market timers. * If a 10% or greater loss occurs from the whole portfolio, go into cash for at least two weeks. Hold any one position’s maximum loss to no greater than 1% of the entire portfolio value. 72

More (Cont) • As the momentum of an up-trend fades, discontinue replacing ‘stopped-out’ positions. More (Cont) • As the momentum of an up-trend fades, discontinue replacing ‘stopped-out’ positions. Let the portfolio extinguish or hedge (d) (see SOTW of 4/13/12) • Go to cash, buy puts, or switch to a bear search, if a follow-through occurs after two red lights (in color guard) (d). 73

Beyond Vector. Vest A consistently profitable trader would also – write and use trading Beyond Vector. Vest A consistently profitable trader would also – write and use trading plans (see next) – use a ‘Stock Selection Guide’ daily (next) – monitor open positions daily and adjust stops as necessary – write a critique after each trade is closed out – maintain an Equity Curve 74

Vector. Vest Trading Plan Bottom Fishing Background Bottom fishing (BF) is the easiest and Vector. Vest Trading Plan Bottom Fishing Background Bottom fishing (BF) is the easiest and most profitable of all possible strategies. Unfortunately, opportunities to trade a true BF only occur 2 or 3 times a year on average. The trend typically lasts 6 weeks and may last longer. Objectives Achieve an average annual gain of 50% or greater during each true bottom fishing campaign. Achieve this gain with a maximum drawdown equal to or less than 20%. Strategies Read the Views, the Color Guard, and Market Timing Graph EOD daily to detect the bottom and the ensuing rally. Complete a Stock Selection Guide daily. Read (listen to) the many SOTWs on BF. Get trading list ready; the turnaround from the bottom comes fast. Use proven BF searches such as S&P 500/RT. Cherry pick search results to select smooth price curves with increasing EPS. Place best prospects in a watchlist. 75

Plan (Cont) Strategies (Cont) Using the watchlist, start to ‘step in’ positions at first Plan (Cont) Strategies (Cont) Using the watchlist, start to ‘step in’ positions at first green light in price column (following a Confirmed Down). Place trades around 10: 30 AM the following day after determining follow-through, using limit orders. Place stop orders on filled positions. As trend gets established, either switch to DEW timer or continue using Green light. Start with G 50/L 10 or 10% trailing stops. Replace ‘stopped out’ positions as long as MACD histogram is rising. When MACD histogram flattens and then decays, do not replace further. Let the portfolio extinguish. As MTI moves up above 1. 4, consider switching out of BF and into another up trending search. 76

 Plan (Cont) Control Routines Enter each trade into a Stock Selection Guide with Plan (Cont) Control Routines Enter each trade into a Stock Selection Guide with description of reason for picking, stop settings, expected result and actual outcome. Monitor the portfolio daily and adjust stops as and if necessary. Prepare and maintain an Equity Curve. Compare results relative to benchmarks like one of VV’s Model Portfolios, i. e. , Yellow Brick Road. A true bottom fishing campaign is defined as entering the upturn, with a BF search, after the MTI is turning up from or near 0. 6 and the BSR is increasing from or near 0. 2. An early indication is when one or more BF searches appear in 1 day Derby Winners. Other BF searches may appear to be better when reading the Derby Winners. They may be used instead. As a general rule, step in only 1 or 2 positions at first. As the trend becomes better established, add more rapidly. 77

Stock Selecting Guide Day: _____ Date: _____ ASSESS (After end of prior trading day): Stock Selecting Guide Day: _____ Date: _____ ASSESS (After end of prior trading day): Color Gauge: _________Price Trend: __________ Color Guard: MTI: ____ BSR: _____ RT: _____Trend ______Call: ____ Views (Brief Narrative) ____________________ Derby Winners (2): _____________________ MTI BSR & RT MACD Trend: _______ Trends: ______Trend: ________ Market condition: Up Off Bottom, Bull Up, Topping, Bear Down Off Top, or Congested. Comments: __________________ Run two searches. Highlight top 30 stocks and Graph all from each search. List best 5 stocks, each with a smooth price curve and rising EPS (for up mkt): List 1: _________ List 2: __________ a. a. b. b. c. c. d. d. e. e. Read stock analysis and recent news (Y/N)___ Final picks: _______________________ Check for near S/R Line (Y/ N)_____Near 52 wk high or low (Y/N)____ Stop setting: _______Target gain setting (if used): ______ 78

Guide (Cont) DECIDE (by 10 to 10: 30 AM): Europe trend and US Futures: Guide (Cont) DECIDE (by 10 to 10: 30 AM): Europe trend and US Futures: _____ Green or Red light in the price column yesterday? (Y or N)____Which? ______ Follow through this morning? (Y/ N) _____ ACT (with broker): Enter trades and stops today? (Y/N) _____ _ If Y, go long or short? _______ Why? __________________________ Fill prices: ________________________ Stops entered (describe): ___________________ OUTCOME -- To be filled in after position(s) is (are) sold: _____________________________________ ADDITIONAL COMMENTS: ____________________________ 79

Lesser Known Features Disclosed • Relating market timers to market conditions • Using stock Lesser Known Features Disclosed • Relating market timers to market conditions • Using stock price timers • Exploiting Value > Price, RS, and/or CI parameters • Cherry-picking stocks • Being aware of and avoiding ‘curve fitting’ • Using MTI and MACD combinations • Separating discretionary from rule-based choices • Benefitting from smart margin use. 80

Credits • A special thanks to Jim Becker, Nancy Nallin, Dan Chisholm and Herb Credits • A special thanks to Jim Becker, Nancy Nallin, Dan Chisholm and Herb Geissler, moderators of Pompano Beach, Sarasota, Orlando, and Pittsburgh Vector. Vest User Groups respectively for inviting the author to present to their groups • Thanks to Ed S. of Orlando, whose eagle eye caught many logic mistakes in an earlier draft • Thanks to Todd Shaffer, Vector. Vest Director of Research, for his input on curve fitting • A personal thanks to Jean, my wife, for her editing. 81

Graph Settings for Cherry-Picking • Initially one year, then 3 months • Price: candles Graph Settings for Cherry-Picking • Initially one year, then 3 months • Price: candles with 200, 50, 20, 10 SMA • Support/Resistance: with 300 bars of data & 3 touches • Volume: with 50 day MA • EPS (look for smooth upward curve) • Value (above price for uptrend) • MACD (default settings of 12, 26, 9) • Opt: Stochastic (default settings of slow, 14, 3).