Statistically Bullish: Crunching Numbers to Beat the Market !

Ever wondered how the Wall Street whizzes seem to have a sixth sense about the stock market? Spoiler alert: It's not magic—it's mathematics. In this blog, we're going to delve deep into the fascinating world where statistics and stock markets intersect, collide, and sometimes even high-five each other. This blog scratches the surface (I say so as this is a exhaustive and debatable topic with every one having their own hypothesis) on what seems to be a logical approach to the equities market.

With a plethora of strategies, tools, and metrics at your disposal, how do you decide the best course of action? The answer lies in a holistic approach that combines technical, fundamental, and statistical analysis.

Fundamental analysis

This is backbone of any long-term investment (3+ yrs) strategy. It helps establish the intrinsic value and overall health of the company by examining its financials, industry position, and economic indicators . So you basically need to dig deep into balance sheets, income statements, and cash flow. This information is readily available online - platforms like Trading View, Bloomberg, Tickertape etc. It scrutinizes metrics like P/E ratios, dividends, and growth rates to gauge a stock's true worth.

Other key factors one looks at are companies with a MOAT or "Rising Giants" in their sector, emerging themes basis the macro-economics or companies operating in a niche sector. In essence, it's about understanding the 'fundamentals' to predict long-term success. Ensure that your portfolio is well-diversified across different sectors and asset classes.

Fundamental analysis is the foundation and the first filter. It helps you filter out companies that align with your investment goals and risk tolerance.

When it comes to establishing the intrinsic value and overall health of a company, several key metrics come into play. Dont get hassled by these financial metrices, there are screener available online which you can customerize as per your requirements. Some of the popular pre-built screeners are Altman Z-score - Solvent/Insolvent screener, Ben Graham screener, Warren Buffer screener, Magic Gormula screener, Pitroski F-score screener each having a different flavour based on these financial metrices.

Here's a rundown:

Some good books to read on this subject are -

Technical analysis

Technical analysis is done by studying candlestick patterns which are nothing but graphical representations of price movements or individual stocks or index etc in financial markets.

Technical analysis is your second filter. It help to time entry and exit points with greater accuracy (helps traders) and to confirm trends or spot potential reversals, adding an extra layer of validation to their strategies (long term investors).

Learning to interpret candlestick patterns is crucial because they offer a real-time snapshot of market psychology, enabling traders and investors to make more informed decisions.

For those new to the game, the basic book to pick up is "Japanese Candlestick Charting Techniques" by Steve Nison. It's the bible for anyone wanting to understand the essence of candlestick charting. Another one is "Candlestick Charting for Dummies" by Russell Rhoads -it breaks down complex patterns into easy-to-understand language.

Some online resources to learn about candlestick patterns are as below :

• Investopedia: Learn more

• WallStreetMojo : Learn more

• Udemy : Learn more

• YouTube: Ultimate Guide

Statistical analysis

Now that you have identified fundamentally strong companies and further done technical analysis by studying candlestick patterns, statistics will help validate these technical patterns and offer additional confidence.

Statistical analysis helps quantify the reliability of technical patterns. For example, if a Bullish Engulfing pattern has an 80% success rate over the past year, it adds weight to your technical analysis. Tools like linear regression can further validate these patterns, offering a more nuanced view of potential price movements.

Descriptive Statistics

These provide a snapshot of past market behavior. Metrics like mean, median, and standard deviation can give you a sense of a stock's historical performance.

Mean (Average): Imagine a stock has prices of $10, $12, and $14 over three days. The mean would be (10 + 12 + 14) / 3 = $12. This gives you a general idea of the stock's value over those days.

Median: If the stock prices over a week were $5, $6, $7, $8, and $9, the median would be $7, the middle value. This is useful for ignoring outliers or extreme values.

Mode: Say a stock hits $20 multiple times in a month more than any other price; $20 is the mode. It indicates a price level where the stock seems comfortable.

Range: If a stock's price fluctuates between $15 and $25 in a week, the range is 25 - 15 = $10. This tells you how volatile the stock is.

Standard Deviation: If a stock usually deviates by $2 from its average price, that's a standard deviation of $2. A higher deviation means more risk but potentially more reward.

Inferential Statistics

These allow you to make predictions about future stock behavior based on past data. Techniques like regression analysis and hypothesis testing can be invaluable here.

Regression Analysis: If every time a company announces increased earnings its stock goes up, you'd use regression to quantify this relationship.

Hypothesis Testing: Suppose you think tech stocks perform better on Mondays. You'd collect data and run tests to confirm or refute this hypothesis.

ANOVA (Analysis of Variance): If you're comparing the average returns of tech, healthcare, and utility stocks, ANOVA helps you determine if the differences in returns are statistically significant.

Chi-Square: Let's say you want to know if stock prices rise more often on Fridays. A Chi-Square test could tell you whether this is random or a real pattern.

Time Series Analysis

Moving Averages : Using a 50-day and 200-day moving average to identify golden/death crosses in Disney's stock.

ARIMA : Forecasting the future prices of Coca-Cola using past price data.

Lets look at some examples to see how technical & statistical analysis work hand in glove.

Example 1: Bullish Engulfing + Linear Regression

Candlestick Pattern: Bullish Engulfing at $50

Statistical Analysis: Linear Regression shows a positive slope

Action: Buy

Entry Point: $51 (just above the engulfing pattern)

Exit Point: $55 (based on regression prediction)

Example 2: Bearish Harami + Standard Deviation

Candlestick Pattern: Bearish Harami at $70

Statistical Analysis: Standard Deviation is high, indicating volatility

Action: Short Sell

Entry Point: $69 (just below the Harami pattern)

Exit Point: $65 (2 standard deviations below the mean)

Example 3: Doji + Probability Analysis

Candlestick Pattern: Doji at $40

Statistical Analysis: 80% probability of price increase after a Doji

Action: Buy

Entry Point: $41 (above Doji high)

Exit Point: $45 (probability analysis shows 10% average increase)

Example 4: Bullish Hammer + Value at Risk (VaR)

Candlestick Pattern: Bullish Hammer at $30

Statistical Analysis: VaR is low, indicating lower risk

Action: Buy

Entry Point: $31 (above Hammer high)

Exit Point: $35 (VaR indicates low risk of significant loss)

Example 5: Shooting Star + Z-score

Candlestick Pattern: Shooting Star at $100

Statistical Analysis: Z-score is 1.8, indicating overbought condition

Action: Sell

Entry Point: $99 (below Shooting Star low)

Exit Point: $95 (Z-score indicates a likely reversion to the mean)

Bonus - Sentiment Analysis (specially for short term investors/traders)

You must have heard that market sentiment is low or high overall or for a sector or maybe for a company. With so much information getting added on the internet, we have a lot of data at our disposal. Text mining by analyzing Twitter feeds to gauge market sentiment could help us tremendously. This can be tricky as social media etc could be manipulated by big players though misinformation or instilling fear unnecessarily to effect price action.

In this intricate world of stock market investing, the blend of Fundamental Analysis, Technical Analysis, and Statistical Analysis can be your guiding star. From scrutinizing balance sheets to decoding candlestick patterns and applying statistical models, each approach offers unique insights. The key is to use them in harmony. Start with fundamental analysis to gauge the intrinsic value and overall health of a company. Then, employ technical analysis to pinpoint your entry and exit strategies. Finally, apply statistical methods to validate your decisions and predict future trends. It's not just about numbers; it's about making numbers work for you.

Happy investing :)

Disclaimer: The information provided in this blog is for educational and informational purposes only. The author has no financial or vested interests in any of the companies, stocks, or investment strategies mentioned. The content is not intended to serve as financial advice or as an endorsement of any particular investment. Investing in the stock market involves risk, and you should consult with a qualified financial advisor before making any investment decisions. The author disclaims any liability for any financial losses or damages incurred as a result of using the information presented herein.