Stocks are random variables

Just to be clear, what I am saying is that the market is not completely random because of the upward bias and the fact that it does trend.

In addition, the market is not predictable because if it were it would die.

In plain English - the market trends, but pullbacks within the trends are unpredictable (random?) and the strength and length of the trends are unpredictable.

Of course that doesn't mean it can't be traded - just don't expect perfection (note to self). Your scaling in and out method makes sense with this in mind.
 
i agree those are reasons, but they are not the only examples of non-randomness

there are (and were) all sorts of edges in the market, in addition to what was mentioned

also note that even EMT people concede the market has an upward bias.

that's irrelevant to whether it is efficient or not (and it isn't - not perfectly efficient)
 
Of course what does "efficient" mean?

Does it mean the market does not react the way one predicted it would? Does efficient mean random or is it an irrelevant term to use in describing the market.

I suppose if the market were really efficient it would flat line as buyers exactly match sellers.
 
as defined (by the ridiculous) efficient market theory - "efficient" means that there is no way to have an edge on the market, without insider information of course.

it is a ludicrous concept, which explains why it is is so beloved among academics (who couldn't trade their way out of a paper bag, and only understand the market to the extent it matches their ivory tower theory)
 
Interesting topic

an "edge" must then mean that the market behaves in a way that one predicts it will - even for a short period of time. With inside information one is predicting that others will find the information bullish. Sometimes good news is used to dump stock so it doens't always work.

A structural edge doesn't require prediction though, like a specialist capturing the spread or having faster communications in the old days.

I guess we can't get away from predicting.
 
There is a huge confusion here because some are interpreting the word random to mean that price is random, and that is not at all what the statistical meaning of random is. In the statistical sense, random means that you observe prices in random order. By contrast, in the market, we do not observe price in anything like random order. Rather we observe price in strict time sequence.

Imagine that you arbitrarly define a population of prices as being all the prices that GM traded at during 2006, for example. You could get that from time and price data. Then you multiply each price that GM traded at during 2006 by the sum-total volume traded at that price throughout the year. That gives you that price's frequency. Then if you plot the frequency on the Y-axis against the prices on the x-axis, you have the distribution of prices for the population of 2006 GM prices. This plot would have 1 tick between each price on the x-axis and the x-axis range would be the total price range that GM traded at during 2006. What on earth that distribution would look like, i have no idea, but would one of you computer geniuses care to make such a plot so we could see? My guess is that it would be unlikely to be even approximately normal and would very likely be skewed, but i'm merely guessing.

Now imagine that you had all that 2006 data in the form of cards where the number of cards was equal to the total number of GM shares traded during 2006. And on on each of the cards was written the price at which each share traded, one price to a card. You shake the cards in a giant hat and start pulling them out RANDOMLY, one at a time, and plot the results, long before you have pulled all the cards out you should have a fairly good estimate of the shape of the 2006 price per share distribution for GM. But you must pull them out in random ORDER , not in time sequence, to be able to do this. And that is the statisical meaning of the word random.

I don't have a clue what this MP stuff is, that you guys are talking about, but it sounds like it must be some kind of distrbution pertaining to the whole market.
 
Quote from whitster:

i agree those are reasons, but they are not the only examples of non-randomness

there are (and were) all sorts of edges in the market, in addition to what was mentioned

also note that even EMT people concede the market has an upward bias.

that's irrelevant to whether it is efficient or not (and it isn't - not perfectly efficient)
More...
People who're struggling with "understand" markets could do a lot worse than simply reading -- and absorbing -- whitster's posts. (And, no, I'm NOT a friend or relative.)
 
Quote from piezoe:

There is a huge confusion here because some are interpreting the word random to mean that price is random, and that is not at all what the statistical meaning of random is. In the statistical sense, random means that you observe prices in random order. By contrast, in the market, we do not observe price in anything like random order. Rather we observe price in strict time sequence.
More...

Thanks, nice explanation. I believe the original poster meant "stock price movements are random." which is how the statement was interpreted.
but
I don't have a clue what this MP stuff is, that you guys are talking about, but it sounds like it must be some kind of distrbution pertaining to the whole market. [/B]
More...

MP=Market Profile. Googling will bring you up to speed.
 
er2. what an edge is, is a methodology that allows you to beat the market.

what that means is not that you have predictive ability, to the extent that i can say with certainty thaT X will happen, but that the predictive potential is great enough such that it will give superior results over time vs. just buying and holding randomly chosen stocks

but an edge doesn't just mean a trading setup like above. it also includes our capability (as daytraders) to NOT trade when conditions aren't optimal

i trade dow index futures.

the majority of the trading day, on the majority of days throughout the year, i am flat (in my trading account. i have longerterm portfolios of course).

i can CHOOSE to enter when the edges develop, and MORE importantly - NOT enter when I do not have an edge.

that is a big advantage, an EDGE, over many institutional money managers who BY CHARTER have to be fully invested in the market

i also have an edge (in stocks and futures) that i am trading small size (relative to GS or MER etc.) such that i can concentrate on stocks that are under the radar (and thus even less efficient) of most big institutions and analysts, etc.

i also have an edge in that i can move in quickly in and out of positions, whereas a big instittuion CANNOT bail on a position (even if they are allowed to), without actually affecting the market (to their detriment).

there are also edges that are based on special knowledge, that do not fall under insider information. computer nerds KNEW (even when the analyst morons didn't) the value of certain game franchises, etc. that certain software companies own

if you know computer games, you can have an edge in trading some sorts of software companies.

if you are in the medical field,. you can have an edge on many others in evaluating new medical products, etc. this is a big part of peter lynch's philosophy - trade what you know, and if you can't explain a stock simply in a sentence or two (why you own it, and what it does), don't own it

again - edges FLUX. because a widely disseminated and utilized edge will necessarily cease to exist. GOOD traders adapt. bad traders rely on the past too much.
 
You must log in or register to reply here.
Back
Top