Can CAPM and APT used for forecasting?

Capital asset pricing model and arbitrage pricing theory are probably the greatest and most popular asset pricing theories of 20th century. They have gone through numerous changes and modifications in the past 50 years since their existence in the 50's. Can they be used to determine asset prices correctly and accurately?

Recently a research has been done and published based on such questions on stock markets of Hong Kong, Japan, Malaysia, Singapore and US. The results shows that CAPM does not even exist in all markets except for Japan as the linear relationship between market beta and asset return can't be accepted statistically. Similar result for APT as well for these markets.

Although CAPM and APT do exist in Japan, their forecasting ability using constant beta and time varying beta are terrible especially during economic crisis of 2007/2008.

So the question is why do we need to study these asset pricing theories? Why CAPM and APT are so popular and basically learned by all finance / business / economic / accounting students? Can we even call them as theories as they are numerous research on similar topics disprove the availability of CAPM and APT in the past.

Using simple economic theory, prices are determined endogenously (provided no government intervention) by market forces of buyers and sellers. The same concept can be applied in financial assets as well. Stocks, bonds, currencies, commodity futures share the same characteristics. Higher the demand, higher the market price, ceteris paribus.

CAPM and APT are not effective tools in predicting and forecating the tradable financial assets (mostly stocks) as they are not used by stock market investors (especially retail investor). Since nobody use CAPM and APT in the decision making process for stock buying, they cannot spur demand and thus, ineffective in such uses.

Most investors (retail or institutional) use technical analysis tools (RSI, Bollinger bands, Stochastic, Macd etc.) and financial analysis tools (PE, ROA, ROE, Profit margin etc.) in determine the buying and selling opportunity in tradable financial assets thus making them much more effective tools in forecasting than CAPM and APT.

So, why don't we introduce technical analysis in tertiary education instead of CAPM and APT? (as far as I know, no university include technical analysis tools in their syllabus, correct me if I'm wrong)?

Or how bout random walk? can asset price be predicted correctly and accurately everytime using certain indicators?

Well, Economists have been discussing this for centuries and answer is yet to be found. Econometricians try to use modelling techniques to form unbelievably complex models and in the end, we still do not have a standardised model in determining asset price. Probably Austrian economists are correct: human being are unpredictable and cannot be estimated using econometric modelling.