top of page

What is Artificial Intelligence modeling of stocks?

Artificial Intelligence (AI) modeling of Asia-Pacific stocks uses a combination of AI, Econometric and Statistical tools  to analyze and provide probabilistic forecasts of trend, support and resistance levels for investors to trade of the stocks of the Asia-Pacific Region. Currently, the markets we cover are: Singapore, Malaysia, Indonesia, Thailand, Hong Kong and India  The models are constructed based on a combination of AI, Econometric and Statistical Models. Inputs: prices data. Output: Probabilistic forecasts of Trend, Support/Resistance prices. Technologies   include Neural Networks, Decision Trees, ARIMA, Boosting/Bagging Ensembles, Monte Carlo Simulation, and MetaLog Distributions.  

  • What are the input and output variables of these models?
    These models can be likened to AI-driven technical analysis models that use only price data i.e. Inputs=Open, High, Low. Output=Close but are minus the many inherent flaws of traditional technical analysis. (TA) What are the inherent flaws of traditional TA indicators e.g. RSI, MACD, Bollinger Bands, Linear Regression etc They all assume that the relationship between market variables is linear and that data distributions are Gaussian (Normal). We know that financial markets exhibit non-linear dynamic characteristics with distributions that are not Normal i.e. have more than 1 peak, are highly skewed and have long fat tails (kurtosis). And that the relationship between market variables is highly non-linear.
  • How then does the methodology of this website overcome the flaws of traditional Technical Analysis?
    There is no one ‘best model’ that fits the modelling of all ETFs. Each ETF’s price and volume data have its own characteristics determined by its type of shareholders, the type of investors who trade it, the industry, sector or market it tracks, whether it is an Active or Passive ETFs, the size of its AUM (Assets Under Management), the expense ratio, the brand name of the issuer and so on. Therefore, we first run all the models in our ‘arsenal’ of tools to find the combination of models that would yield the best performance for that ETF. AI models like Neural Nets, Decision Trees and Random Forests handle non-linearity easily through their activation functions, nodes and hidden layers of neurons. Econometric models such as ARIMA and Holt-Winters Smoothing help to distinguish between noise and true signals Decision Trees and Random Forests of Decision Trees are used for data where Neural Networks tend to overfit. Ensembles harness the output of multiple copies for a consensus that reduce the margin of error. Statistical Bagging (random sampling with replacement) and Boosting (enhanced gradient descent) ensembles increase model performance. Monte Carlo simulations test the model by running 1000 trials and establishing the Probability Density Functions (PDF) Results are fitted to MetaLog PDFs which are best suited to are financial markets data to establish the 5% and 95% Confidence Levels for Support and Resistance prices .
  • What can these models do?
    They can cluster and classify ETFs into groups with potential for arbitrage, forecast the probability of prices for a specified period in the future, and create ETF portfolios that are optimised for the financial objective of the investor. However, for now, the models on this website will only be used for establishing the probabilistic Support and Resistance price levels of an ETF. The default probabilistic Support and Resistance are at the 5% levels i.e.” There is a less than 5% chance that the price will be lower than $x and 5% chance that the price will higher than $y.” However, they can be adjusted, by request to the investor’s choice of risk level % Investors can use the establishment of the range between Support and Resistance to guide their trading.
  • Model Output Report
    The Report will be in two parts. Part 1 consists of 3 charts: (a) the trend of the prediction 1-20 days ahead (b) The fitted MetaLog Distribution and the Support and Resistance price (c) Q-Q chart showing the degree of fit of our simulation output. Followed by an analysis of these charts Interpretation of these charts will be explained in the first article to be posted in The Learning Corner. Part 2 will give the details of the ETF being written about: AUM, average daily trading volume, issuer, brand, date of inception, analyst report

Why the Asia-Pacific Region?

There is no doubt that the 21st Century belongs to Asia. Especially China, India, and the 10 countries of ASEAN (Singapore, Malaysia, Indonesia, Thailand, Vietnam, The Philippines, Cambodia, Laos, and Brunei). With a combined population of 700 million, ASEAN will ride on the tailwinds of economic giants India (1.4 billion population) and China (1.4 billion population). The region will in the years to come contribute an increasing share of global GDP.  Intra-ASEAN cross border trade is also thriving, and ASEAN countries with large and young populations like Vietnam (100 million), The Philippines, and Indonesia (270 million)  have the huge domestic consumer demand that adds a degree of resilience to their economies besides driving intra-ASEAN trade and investment.  While China stocks are not readily accessible to retail investors, the Hong Kong Exchange with its many mainland  companies is a good proxy. The sheer dynamics of the Asia-Pacific Region's  demographics, its innovativeness, entrepreneurship, combined with rich natural resources, good transportation, digital communication and financial infrastructure (Singapore and Hong Kong) makes the realization of its potential inevitable 

LATEST POSTS

bottom of page