TECHNICAL INSIGHTS

The Whiskystats Indices

16. October 2016 | 2K Views
Please note that this article is based on the old Whiskystats database.
Some numbers and figures may not match what you find in the Whiskystats 2.0 database.

We here want to explain how we calculate our indices, why we decided to do so and what other options there might be. This should give you the possibility to fully understand and correctly interpret the results we come up with. 

In our price update of August 2016 the index value of our Whiskystats Whisky Index (WWI) was below 300 points and we wondered if it will soon jump above this threshold. One month later, in September 2016, we observed that in the updated index the value for August 2016 was suddenly above 300 and dropped to 296 index points in September 2016. How is this possible?
 
Current Method
 
The intention of our indices is to summarize big portions of the single malt whisky market (or the whole market like in the WWI). This helps us to get an overview of the big market movements. For this purpose we want to use all the information that is currently available so that we get the most accurate picture.
 
For the WWI for example, each month we determine the all-time 300 most traded (*) single malt whiskies. We then summarize the price development of this virtual collection with one index, which always starts with an initial value of 100. So the index simply describes how the value of these 300 whiskies changed over time. This easy and straight forward interpretation is the reason why we chose this method.
 
The drawback of this method is that the index is not consistent over time. As the set of included whiskies dynamically changes each month and since we recalculate the whole index history it is possible that those historic index values will not be the same next month. For us, this is totally fine since this method delivers the most information possible given the current data and we exactly know how to interpret those numbers.
 
Alternative Method
 
There are, of course, ways to ensure consistency over time. One possibility would be to fix the set of included whiskies. But this would not capture the dynamics of the market and we do not want to artificially decide which are the most important whiskies to consider in the index.
 
Considering this, another possibility would be to determine the latest price change of the current set of most traded whiskies and multiply it to last month´s index value. Doing so, the historic index values stay the same by simultaneously capturing the dynamics of the market, i.e. always considering the most traded whiskies based on the current data.
 
The major drawback would be that there is no straight-forward interpretation of this index´ values. Many different whiskies would have contributed to it. Some may did so for only a short period of time. And what if we expand our database to also include Bourbon (for example)? How would we squeeze those historic prices into the WWI?
 
Consistency vs. Interpretability
 
So in the end it is a decision between interpretability and consistency. We either have an index for which we know exactly what it describes, or an index which historic values do not change. Since our goal is to bring more transparency to the secondary whisky market, we initially opted for the easy interpretation. And maybe there is a method to combine both, transparency and consistency. We haven’t found it yet, but would be happy to adapt it if you tell us how to do so!
 
If you need further information on our indices do not hesitate to leave a comment. Also, if you know a better method that combines consistency and interpretability we would be happy to discuss it and, if possible, adapt it!
 
(*) In the context of index calculation we use the number of different auctions a whisky appeared in rather than the sheer number of traded bottles to determine how much a whisky is traded.