Newbie Lessons - All your questions answered

[Sir Osisofliver]

So I'm going to continue with Fundamental Analysis for a bit and then move into technical analysis. (What's the point of having such a large heading with only a paragraph behind it eh?)

By no means have I used an exhaustive list of metrics that can be calculated for fundamental analysis, there are plenty more out there. A common feature of ALL metrics used in fundamental analysis however is that they are all used for comparative purposes. It's important to realize this about fundamental analysis. It's only worthwhile using these metrics if you can compare apples to apples. So using P/E ratio on CBA, and comparing it to CBA two years ago...great...apples for apples comparison. Using the P/E ratio for CBA and comparing it to the P/E Ratio for NAN...completely useless and pointless. The object of comparison needs to have a close relationship for the metrics to have any relevance.

One of the things you may have heard mentioned is the concept of Top Down research or Bottom Up research. Because there are so many comparatives out there for fundamental analysis that can be used how do we get meaningful data out of noise, and how do we decide where to look? Analysts generally use Top Down and Bottom up approaches.

Top down looks at the macroeconomic picture first and attempts to identify from the prevailing economic conditions which sectors of the market will outperform. From this pespective it will then look at what macro factors each stock within that industry have, eg barriers to entry, intellectual property things that differentiate the stock and make it more attractive to a greater range of investors. (and therefore give us a higher probability of a beneficial outcome)

An example of a top down approach using Fundamental analysis is for example the CBM companies in the energy sector. By looking at the fundamental macro effects we can see why larger international companies are being predatory towards our CBM players at present. They are not her to service the 25 Million people that live in Australia. They are comng here to service the 2 billion people that live in Asia over the next ten years. Macro effects

Bottom up approach is the analysis you do using the metrics I mentioned above, and include things like what does the balance sheet look like, what forecasts are in place etc etc. It's all about the numbers and how they compare with similar companies and whether those numbers indicate "value".

OK I lied, Technical Analysis will be coming in part three (Son of Tech/A)

Cheers

Sir O

 

[Sir Osisofliver]

Technical Analysis

OK I'm finding myself with somewhat of a daunting task here. There are a number of different techniques that are used in technical analysis. I don't think there is any way that I can cover off against all of them and compress all of that information into a simple to understand and easy to interpret summary for beginners in a few pages of text (and a few charts if my Market Analyst ever bloody works again - damn MA6 razza frazza). I don't plan to put myself forwards and blow smoke up my own back end to say that I'm an expert of all styles, so what I cover here will be general in nature. Perhaps when I get to part two (not so newbie lessons) we can have some of the experienced members of this forum put forward the systems and techniques that they use to achieve their aims.

Technical analysis and systems to exploit the results of that analysis with consistent results takes years to accomplish - if ever - some people never achieve a consistent system. I've been doing technical analysis for a while now and I'm still learning and will probably do so until they put me in the ground. What may work and be extremely successful for one person may fail abysmally for the next person. (It's a bit like Schrodingers Cat for the stock market - your OWN emotional state can have an effect on the way that you perceive the information you are trying to extract.)

Largely this is because Technical Analysis seeks to quantify emotion in the marketplace, and how one person feels about an issue may be completely different to what someone else believes. Just have a look at the Housing Market threads. https://www.aussiestockforums.com/forums/showthread.php?t=4452

Or the dead cat bounce thread... https://www.aussiestockforums.com/forums/showthread.php?t=14689&highlight=DCB

Ask a question like "have we bottomed yet?" or "Are Interest Rates going to fall further?" or "Will the American economy stabilize in the next 3,6 or 9 months?" and you'll inevitably find that on the issue you will have rabid and emphatic supporters of the concept who will pull up stats to support their view. You'll ALSO get rabid and emphatic detractors who will say the exact opposite. You'll ALSO get those people who don't have a clue either way and are reactive to whatever the prevailing wind says and the people who believe something somewhere along that range.

With the use of technical analysis you're trying to capture and exploit the aggregate result of all that emotion primarily by the use of patterns and trends and also not have your own emotional bias skew your consistency of getting the probability of a beneficial outcome.


Sorry Newbs - the wife is giving me "the look" and dinner is on the table - I might be back later to continue this.

Sir O

 

[Sir Osisofliver]

The ultimate aim of Technical Analysis is to accurately predict the movement of a share price based solely upon the price and volume data that has already happened.

It's use by many academics in the field of Finance is viewed as dubious and inconclusive at best, and merely a revenue generating exercise for broking houses at worst. The theory of EMH is double edged. If someone finds a consistent and predictable way to use analysis (either technical or fundamental) the moment that the market becomes widely aware of this methodology, system or indicator, it will be incorporated by the market and the advantage will disappear. So theoretically if you find something...keep it to yourself until you've made your billions.

In all likelihood in my opinion since we are dealing with complex chaotic system analysis, no one will ever find a methodology or indicator that will get it right every time. This is why I talk about raising your probability of a beneficial outcome, because we are trying to use imperfect mathematical tools to analyze a complex chaotic system. The best we can hope for is to raise our probability of returns and get it right more often than we get it wrong. If you want to learn more about Chaos Theory there are some interesting articles around if you search for them but most of them will do your head in. (This is also why I use fundamental analysis - to further raise my probability of a beneficial outcome).

I'm telling you this not because I want to discourage you from using technical analysis but because I want to express how difficult it is to achieve a consistent and profitable system. If someone tells you. "Look it's simple you just need to use a wave analysis Gann, Fibonacci, EW approach, Candlestick with a stochastic oscillator and a MACD, Bollinger Bands, and look for H&S, Double Tops, Resistance Levels, Flags, Pennants, Triangles, and Dodecahedons (that last one is a joke BTW) - they are full of it. By all means examine what others use to achieve success, by all means create a system based upon what others have done and see if you can make it work for yourself, if you find something that works - use it. If you can't make it work....discard it and move on and their is no substitute for experience.

Now for some good news... There are some who are so successful that they have amassed huge fortunes from the use of technical analysis. These people like Larry Hite and Ed Seykota, William Eckhart, John W Henry, Victor Sperando and George Lane (He's the guy who coined the phrase "the trend is your friend") by their very consistency and success indicate that it is possible to use technical analysis with great efficiency. By their very existence and track records they challenge the theory of EMH.

Ok next episode I'll actually begin discussing and giving examples on how to use some individual tools (If my MA6 is working by then and I can whack a chart up here).

Cheers

Sir O

 

[prawn_86]

Fantastic stuff yet again Sir O, keep up the good work

 

[Sir Osisofliver]

Hi All,

Just a quick note - I'll be snowed under over the next couple of days. Don't expect any updates till next month.

Sir O

 

[Holy Roly]

Sir O....

Where do you download your data from?

How much does it cost, if at all?

What software do you use to run the data through?

and, more personally...

Are you amassing anything close to a fortune?


Note: just trying to buy into TA trading, but still not convinced!

 

[Sir Osisofliver]

Sir O....

Where do you download your data from?

Fundamental or technical? Market data is Paritech

How much does it cost, if at all?

I don't pay for it (the firm does)

What software do you use to run the data through?

Market Analyst software

and, more personally...

Are you amassing anything close to a fortune?

I don't yet have a solid gold toilet seat or a private jet

Note: just trying to buy into TA trading, but still not convinced!

Holy Roly How are your math skills? - This is a newbie thread so I won't go into huge amounts of detail here. You can see above what a large number of people still think about the use of technical analysis. That TA is all so much BS and fundamentals are everything. When I went through University, the same tired old concepts of fundamental analysis were taught to me (and they still are), all based upon random walks and probability analysis. You may even see that Standard Deviations are still used to describe the probability of events within a marketplace and lets not forget the good old Bell curve style of probability analysis. All this type of analysis is based upon "pure" probability analysis. IE Toss a coin 100 times and pure probability says that it will land 50 times heads up and 50 times tails up.

Totally random results, no interconnectedness. But does it actually describe what happens in the market? Care to guess what the probability is that our market will drop 25% overnight like '87? It's about 10 to the power of 50. That's a lot of zero's eh? That's the sort of number that gets used to describe the number of planetary bodies in the galaxy. So much for totally random with no interconnectedness eh? So Pure probability and random walks understate the risks and cyclical nature of markets.

Why does it get used then? It's a tool that is "good enough" about 85% of the time. So during that 85% they get to make hay while the sun shines, and during the 15% where the market tanks? They get to blame things like black swans and say they couldn't see it coming.

This is why I discuss chaotic systems analysis, because it's a better model for prediction of outlying probabilistic events on bell curves and increasing your probability of a beneficial outcome. If you want to learn more about chaotic systems analysis, go read "The (Mis)behaviour of Markets" by Benoit Mandelbrot. It's an interesting read and not too technical.

Cheers

Sir O

 

[Sir Osisofliver]

A bit on technical analysis.

OK so I'm going to talk about the types of Technical Analysis that I employ and the reasons why. Fair warning - this may challenge some of you who are using tools different to myself. I'm going to say what I use and WHY I use it - and I do so relatively successfully. But to be fair - my way works for me. It's not going to work for everyone. My advice to anyone trying technical analysis is to try ALL styles - abandon those that do not work for you and hone your skills in those that do.

Ok so if you look in the post above you can see what I think of traditional risk evaluation methods. I don't really mind that almost every analyst and advisor out there is trained in classical methods. It gives me an edge over them that I have exploited for my own benefit time and time again. I don't get it right all the time, but I get it right a much larger percentage than I get it wrong. If you disagree with what I am about to say - please engage - I ALWAYS like challenges to my methodology - it makes me better at what I love to do.

So lets think about risk... It's a really cool question to consider...how do we measure risk? You look at a ahcart and it goes up and down and as a pattern seeking animal with wish fulfillment you start a game where you go...if I bought here and sold here...... and then try an think about the risk involved. If you are a firey french mathmatician you use variance and standard deviation as a proxy for risk. Standard deviation is the mathmatical base behind a very large number of finance tools, both fundamental and technical. These have been honed and built upon over a large period of time and given rise to Nobel prize-winning concepts like the Efficient Market Hypothesis, Capital Asset Pricing Model and Modern Portfolio Theory among others. (by the way if you are curious and want a chuckle - and also learn something that they won't tell you about Harry Markowitz - go research what happened when Markowitz decided to try reality rather than academia - you'll need to search for it as embarrassig black marks when you are the Father of Modern Portfolio Theory get obscured but it makes for very interesting reading).

It's these tools that you or I as investors try and tease alpha out of the market. I don't know how to slot a picture in right here, but below there will be a picture of a bell curve, with standard deviations marked. This is how a very large percentage of the worlds brokers, planners, fund managers and analysts measure risk - on an individual stock - or across a portfolio.

Once again this is my opinion - but standard deviation as a measure of risk IS BROKEN. How can I say that? What's the odds and standard deviation that a market (and hence a portfolio) will lose 29% of it's value in a single day? The answer is in the above post 1:10 to the power of 50 or over 20 standard deviations. Look at that bell curve...where is 20 standard deviations? It's about three pages to the left of where that chart ends. It's so far outside of the bell curve that it should never have happend if the bell curve was an accurate assessment of risk. Bah! the '87 crash was an anomoly I hear you say? Ok exclude it. What's the math behind a simple 7% movement of the market? Standard deviation tells us that it should happen very very rarely. (sorry I don't have the numbers handy) In reality it happens about 1500 times more frequently than the theory suggests. (I'll also say that figure comes from my imperfect memory - I can't be arsed to dig up the figure - either trust me that the statistical frequency of outlying probabilistic events in the market are much more frequent than the theory behind it suggests by a huge margin - or research it yourself).

What's the purpose of this little rant? To get you to look at what tools you use and figure out which ones are based on faulty math. Go look up your favourite indicators - Do you use a Bollinger Band? (which incorporates a simple moving average with a standard deviation) - How about a MACD? Stochastic Oscillator? How about a Relative Strength Index? these a just a few of the many many indicators with Standard Deviation and variance at their core. Why do these things still get used if they are broken? Because they are right around 85% of the time [the inner part of the bell curve] (which you can still make money from with proper risk management in place). By all means use these indicators - but I've found that having an awareness of their limitations and understanding the math behind them is invaluable.

Sorry guys - this is taking longer than I thought - the mystery of what I use will need to wait for another day - too much to do.

Cheers

Sir O

 

[Sir Osisofliver]

hopefully the picture is attached now (I can't tell)

If It isn't http://en.wikipedia.org/wiki/Standard_deviation

 

[prawn_86]

Yeh, we have been taught standard normal distributions at uni. It sucks, cause i have to know it and use it for assignments, despite the fact that it is just utter cr@p in my opinion.

Academia is always behind the curve (pun intended ) anyway

 

[Sir Osisofliver]

Yeh, we have been taught standard normal distributions at uni. It sucks, cause i have to know it and use it for assignments, despite the fact that it is just utter cr@p in my opinion.

Academia is always behind the curve (pun intended ) anyway

Prawn - less than five years ago I was doing post grad stuff on advanced portfolio management (the stuff you do to become a fund manager) - and this stuff was being held up as the be all and end of of risk assessment.. The uncomfortable fact for academics in this field however is....

If you take it away....what do you replace it with? How do you measure risk without variance and standard deviation?

Until a better theory rises to provable prominance the old theory will be defended. Some people will look at what I have said here and say.."That Sir O! what a freaking cheeky sod, Markowitz won the Nobel prize for economics, where's Sir O's Nobel prize eh? Yeah he's just talking rubbish and taking pot shots with nothing to substantiate what he's said but anomolies in the data stream. It works 85% of the time!! (Of course I view this behaviour as running around in circles shouting the earth is flat - maybe that is just me.)

Cheers

Sir O

 

[beamstas]

Thought i'd offer some advise here as no one else is

Where do you download your data from?

Premium Data. Premiumdata.net. You really get what you pay for when buying data, if you want free data it will normally be crap. Premiumdata is EOD data only but you can get intra day snapshots for about 20cents extra a day.

How much does it cost, if at all?

$30 per month + $5.50 per month extra for intra day snapshots. That is asx only data, they have a pricing table take a look at premiumdata.net.

What software do you use to run the data through?

Amibroker. That is imo the best for someone getting a "feel" for using Software. It has everything you'll need to get started, including pretty good graphing capabilities, backtesting (testing systems automatically from historical data), and pretty much everything else you'll ever need

Note: just trying to buy into TA trading, but still not convinced!

There is a big difference between T/A and Fundamentals, it's like choosing between Ford & Holden Personally i use T/A as i do not believe fundamentals to be accurate. You'll have to make your own call on this!


Both Amibroker and Premiumdata offer free trials so you can determine if they are what you want

Brad

 

[Bat_Ears]

Thanks for posting this Sir O. I am looking forward to the next entry.

 

[Gordon Gekko]

Thanks for posting this Sir O. I am looking forward to the next entry.

Once again thanks for all the great info! I'm hoping a future lesson will involve options!
I would like to learn how to protect my investments without having to sell them directly. Or have the option to buy.

Best

G

 

[MRC & Co]

Once again thanks for all the great info! I'm hoping a future lesson will involve options!
I would like to learn how to protect my investments without having to sell them directly. Or have the option to buy.

Best

G

WayneL had a great blog on options, you won't get any better than it (no offence Sir O)!

 

[Gordon Gekko]

WayneL had a great blog on options, you won't get any better than it (no offence Sir O)!

Thanks for that MRC,

How do I find it here on the sight?
I tried to search for options but it was to broad? Maybe search for WayneL posts?

Thanks and all the best

G

 

[MRC & Co]

Thanks for that MRC,

How do I find it here on the sight?
I tried to search for options but it was to broad? Maybe search for WayneL posts?

Thanks and all the best

G

Yep, threads created by Wayne L and options as a keyword.

Also, go to his profile page, he used to have a link down the bottom to his own actual blog on options, it's a step by step introduction and guide to options.

A few others around here, such as Sails, Mazza and Magdoran (probably missed a few, sorry), are skilled in the area also, so will happily answer any of your questions.

Just takes a while to get your head around the concepts, but after that, it's not overly complicated unless you are going to use them as your main instrument, then you would want to be extremelly proficient, which would take years.

Good luck.

 

[Gordon Gekko]

Yep, threads created by Wayne L and options as a keyword.

Also, go to his profile page, he used to have a link down the bottom to his own actual blog on options, it's a step by step introduction and guide to options.

A few others around here, such as Sails, Mazza and Magdoran (probably missed a few, sorry), are skilled in the area also, so will happily answer any of your questions.

Just takes a while to get your head around the concepts, but after that, it's not overly complicated unless you are going to use them as your main instrument, then you would want to be extremelly proficient, which would take years.

Good luck.

Ive just been reading his posts! Thanks again.

Seems I've had a bit to much wine to take it in at the moment so I'll have a look tomorrow.

Thanks again

G

 

[MRC & Co]

No problemos Gordon.

Wine, mmmm, might pop a bottle myself!

 

[tech/a]

Once again this is my opinion - but standard deviation as a measure of risk IS BROKEN. How can I say that? What's the odds and standard deviation that a market (and hence a portfolio) will lose 29% of it's value in a single day? The answer is in the above post 1:10 to the power of 50 or over 20 standard deviations. Look at that bell curve...where is 20 standard deviations? It's about three pages to the left of where that chart ends. It's so far outside of the bell curve that it should never have happend if the bell curve was an accurate assessment of risk. Bah! the '87 crash was an anomoly I hear you say? Ok exclude it. What's the math behind a simple 7% movement of the market? Standard deviation tells us that it should happen very very rarely. (sorry I don't have the numbers handy) In reality it happens about 1500 times more frequently than the theory suggests. (I'll also say that figure comes from my imperfect memory - I can't be arsed to dig up the figure - either trust me that the statistical frequency of outlying probabilistic events in the market are much more frequent than the theory behind it suggests by a huge margin - or research it yourself).

And in your view which form of analysis or indeed investment mitigates this outlier risk factor?