Apologies for the delayed response. I wholeheartedly agree we shouldn't condemn any approach, and certainly not anyone who follows an approach if it makes sense to them.
TA purports to be able to predict short term price movements. There would definitely be a use for any method that can do this (even inaccurately). However of the TA approaches I am familiar with I am not convinced that any of them are logically sound.
Let's have a look at the logic.
Input data for TA:
* Prices of previously conducted trades
* Volume of trade (including changes in vol)
* Existing orders on the market
Also Considered:
* Amount of short selling
* Dividend payments and relevant dates (probably)
* Reporting dates (possibly)
Let me know if i'm missing anything. My understanding is that TA focuses mainly on the first 3 of the above inputs. We will exclude fundamental data since that isn't used in TA.
Desired output:
* Future price
TA Theory:
* we can divine market psychology based on the above inputs to get an idea of the future price.
Here's some thoughts on the above... a bit off the cuff and I'm happy to have any of this contested:
For determining future price, the prices of past trades can indicate roughly what people perceive (correctly or incorrectly) to be fair value over long term in light of non-technical indicators, but I can't see that it tells us much about what will happen in the short term.
Volume increases/decreases indicate changes in level of interest in the stock which impacts liquidity but not much else. Very low levels of liquidity (e.g. with penny stocks) can lead to wild jumps in price , but past a certain level it makes less and less difference.
Existing market orders can be changed at whim prior to being closed. Some are placed specifically to try and manipulate price (whether or not this actually works). This might tell us something about 'resistance' on large price swings if the price moves too quickly for people to remove their orders. But those with active orders are usually monitoring them closely (that's before we talk about automated trading).
Of the "also considered" points above, these all have valid impacts on price that are semi predictable. But I don't see that the reporting/dividend dates are of much practical use. Maybe the impact of shorting can be inaccurately predicted (and therefore arbitraged) but that's about it from what I can see.
SHJ Price at posting:
$1.23 Sentiment: Buy Disclosure: Held
