emh implications: technicians use information from past prices to forecast future price movemements. this violates the random walk model of stock prices and the weak form of the emh. emh says that the market price already contains information in past prices. random walk says prices are generated by a weiner process, i.e., price movements are independent.

empirical evidence: the empirical and anecdotal evidence is consistent with the emh. the evidence shows that chartists are not able to beat the market on a consistent basis. many such papers are listed in my annotated bibliography. the evidence against the emh consists of some known anomalies. examples: the january effect, the weekend effect, nfl effect, and the election effect.

objective of technical analysis: market indicators and chart patterns trigger trading rules (also called "filter rules"). these trading rules are expected to outperform emh strategies. (emh strategy = buy and hold the index)

technical analysis in practice: technical analysts respond to the emh argument in three ways. 1. newer ("more sophisticated") analysis tools. 2. claim that markets are efficient because they are actively removing inefficient pricing, 3. avoid purely technical trading rules. many asian newspapers (bangkok post) publish technical analysis information daily.

number of issues advancing - number of issues declining.
applies to market, sector, or industry but not to a single security
large positive number: bull indicator for the "broad market"
large negative number: bear indicator
used to confirm index movement. if not confirmed, the market is called "mixed"
large positive value in conjunction with large increase in index: very bullish. market is said to show "momentum"
extension of this logic: trin statistic = avg volume of declining stocks / avg volume of advancing stocks
trin greater than unity considered bearish

short interest outstanding: short sales that have not been covered.

if the number is large then we know that these short sellers will be in the market buying, so it is a bull indicator
or: if short sellers have better information than the market at large, then short interest is a bear indicator

cash position of stock funds

stock funds are limited in their ability to hold cash by the prospectus
increasing cash position: places fund managers in a "have to buy" position so it is a bullish indicator
or: if fund managers have better information, it is bearish because "informed traders" are out of the market
credit balance in broker accounts have a similar interpretation

investor confidence index

confidence index = yield spread expressed as a ratio of high grade/low grade
when investors are optimistic ratio approaches unity or 100%
rising confidence index is bullish indicator for stocks
the index is published by barrons

put/call ratio

number of outstanding call option contracts / number of outstanding put option contracts
normally around 60%-70%
rising ratio out of this range is bullish and falling ratio is bearish if option buyers have information
the opposite is true if you believe that option writers have information

odd lot trading

theory: odd lot traders are small investors and do not have information
odd lot traders miss market timing points (they are late)
sudden increase in odd lots is a bearish indicator

relative strength

track stock price as ratio of P/index
rising RS is bullish

the valueline strategy

value line ratings: 1 (best) thru 5 (worst) forecast stock performance over the next 12 months for 1700 stocks followed by valueline
empirical evidence shows that valueline #1 group outperforms the sp500 and the #5 group underperforms the sp500.
valueline uses relative earnings growth and short term and long term average relative strength to compute their ratings.

dow theory
market trends: primary, secondary, minor: bull/bear
resistance and support levels

congestion and breakout: long shallow bar followed by move out of the bar is a breakout. direction of breakout signals bull or bear

fulcrum: breakout downward more dramatic if congestion preceded by steady decline. in p&f chart, a vertical drop followed by congestion