1. Getting Started: The Research Portal

z-score dashboard

Performance Table

The name of the site says it all: Signal 2 Noise.

There is so much noise in the financial markets, amplified through TV, print, and social media, that focusing on what is important is really hard to do. The main objective of this site is to provide a way for people to consume as much relevant information as I can produce in a fun, easy-to-understand way.

Start with an Overview

I believe the first thing you need is a table that produces returns over different time periods in one snapshot. The table below is my starting point.

I provide the symbol ticker from Yahoo Finance, a free resource, and the full name description. As you can see, I included the price. I think it is always good to have a feel for the actual price of the symbol, but the most important information is the % return. I stay away from the point change, as it can be very deceptive. A 1,000-point move on the Dow sounds like a lot, but it is only a 2% move in the current market.

The next feature in the table that makes it different from most is the colour coding, which highlights a signal.

There are two main problems when looking at % returns:

  1. The first issue is not knowing whether the move over a particular time period is a big or small move relative to its history. For example, the EURUSD moved 1% over the previous day. You think 1% is not a big deal, but in fact, a 1% move for EURUSD is a relatively big move.
  2. When you look at % returns in relation to other symbols, you cannot judge a 1% return as the same for all symbols. For example, a 1% move with EURUSD may be considered a big move, but a 1% move with Bitcoin is average.

 

Z-Score Signal

To overcome these issues, among a few others, we use the Z-Score to highlight whether there is a % return that is worth highlighting, what I call a signal.

The Z-Score formula = (Return – Mean) / Standard Deviation 

The parameters I have used for the calculation of the table are as follows:

  • Lookback period for the calculation of the mean and standard deviation is 5 years.
  • + 0.5 Z-Score is light green.
  • – 0.5 Z-Score is light coral. 
  • + 1 Z-Score is dark green
  • – 1 Z-Score is dark red.

It is important to remember to make sure day returns are matched with mean daily returns and daily standard deviation of the lookback period.

Best and Worst YTD Performance

This is a very simple visualisation using horizontal bars to express the YTD performance from best to worst.

Price Journey YTD Performance

I find this visualisation one of the most powerful ways of showing performance. The table at the top and the horizontal bar above show the current value, so it is a snapshot of one point in time. With a time series of the performance, you can see the actual journey. The green shows if the current % performance is positive, or red if it is negative.

Rolling Volatility

I find this visualisation one of the most powerful ways of showing performance. The table at the top and the horizontal bar above show the current value, so it is a snapshot of one point in time. With a time series of the performance, you can see the actual journey. The green shows if the current % performance is positive, or red if it is negative.

Rolling & Matrix Correlation

The same principles apply with the rolling correlation. In this example below we have picked the SP500 (^GSPC) as the base symbol and then we have applied a number of different symbols against the base symbol to see the 1yr rolling correlation over the last 10yrs. Once again we see that correlations can be highly variable through time.

We also present a correlation matrix with the data going back all the way to year 2000.