**Introduction**

Almost four years ago I carried out an analysis of two models of second hand cars then on sale in Bangkok. This was a follow up to something I had done before for some cars in the UK. That page is here, on this blog. I wrote a follow up page a while later entitled **Depreciation Schedules: BMW, Volvo and Porsche**; again on this blog, here.

**Depreciation Schedule Estimation: Bangkok and Melbourne**

For this page I have returned to Bangkok and found the prices of 64 second hand cars: **Toyota Vios 1.5 (E)**. The prices cover the period 2003 to 2014 and I copied them from the web site market doorot. I also took a look at the data for the **Toyota Yaris Ascent** for Melbourne in Australia covering the period 2014 to 2016.

For Bangkok I only got age and price data: I got gear box data too but all but two of the cars on sale were automatic so that distinction was not useful.

For Melbourne, they provided only three years’ worth of data which limits the analysis but they usefully provided kilometres driven, so that is my independent variable in that case. They also provided gear box details and there are more manual transmission cars than in Bangkok and that is useful in this case. I took the Melbourne data from this site.

**Some Results**

The screenshots below are taken from my working file on this update and they show the pivot tables I produced together with their pivot charts.

**Bangkok**

**Melbourne**

The Vios update from Bangkok shows a very interesting change: four years ago the rate of depreciation of the equivalent car was about 3.8% per year; this has risen to 5.3% per year in my latest analysis. Thus, estimated useful lives have changed from 26.4 years to 18.9 years.

Let me stress, as I did before, this is all based on informal analysis and I have made no attempt to carry out any fully scientific analysis involving control groups, random assignment of cars to control groups and so on. In this latest update, I just went to the first site I found and took all of my data from there: that alone, of course, could explain the changes in depreciation rates and useful lives. However, I am sure my data are indicative of the general situation in Bangkok where cars have historically held their value.

As for my Melbourne data, I have never analysed any Australian vehicle data before so this is new to me. Moreover, my depreciation model for Melbourne is a kilometre driven model and not a time series model. My model is summarised in the screenshot below:

At this stage I have not carried out a fully regression analysis on this update and my Excel file is available **only on demand**: you ask, you get!

Duncan Williamson

9th November 2016