I decided to do my next analysis on avocado’s. Why? Just because I LOVE avocado’s and since avocado’s suddenly are seen as superfood they seem to become more expensive every year, and I want to know more about that. I want to see that in a chart, I want to see numbers. I randomly bumped into an avocado data set, and did I already say that I love avocodo’s? Before I start, I’ll say this is not the most complete dataset of avocado information worldwide. It seems to be a dataset with US regions specifically, but since avocado’s grow world wide so this dataset is definitely not complete.
In this article I will focus on a few interesting variables in the dataset. Price, region, type of avocado (organic vs. conventional), date, years & weeks to see if there is significant differences between variables and if one affects the other. After checking and cleaning up (incomplete or unnecessary) data, I transformed the date into weeks and years, to be able to compare the average price per week, over the 4 years in the data set (2015 – 2018).
Average price per avocado over 4 years (2015 – 2018)
Looking more closely to this chart, something interesting is going on.
2015
The black line, (2015) is not showing strong differences throughout the year. There is a bit of fluctuation with a dip at the end of the year, but that’s about it.
2016
The purple line (2016) is starting to show something slightly different, there is a dip in the average price around week 13-18, and a peek in the price around week 39 – 43.
2017
The blue line (2017) is showing interesting insights in a whole different way again; there is a dip early in the year. 2017 seems to be an almost ongoing increase in average price per avocado per week. It reaches a high around week 35 up to 39 and decreases quickly after. Back to the same level as where it started.
All three years, 2015, 2016 and 2017 show a dip at the beginning of their year, around the same time. It is a dip that only lasts for one week and the price goes up after. other than that, I left 2018 with a white line, since it only has data for the first three months, which does not give a good overview or indication for the rest of the year.
Price per week per type
Looking at the broad overview of differences in price per type of avocado (conventional or organic) throughout the year, we’ll see a very strong first conclusion that organic avocado are (as expected) more expensive all year around. The price difference seems to be stable, except for the dip in the beginning of the year. That price dip only seems to be the case in conventional avocado’s and not the organic ones. Apart from that the price moves more or less synchrone with a peek around week 37-39. The price for both conventional and organic avocado’s hit rock bottom at the end of the year.
Interesting insights already, but let’s go one step further.
Price per week per type, split per year
Below four different charts that show the progress of the average price per avocado per week, split per year.
2015
The average price of conventional avocado’s in 2015 are pretty stable throughout the year, between $1.00 and $1.20. No massive dips or peeks at all. The organic ones are more expensive throughout the year. and there is a little peak between week 30 and 40.
2016
The average price of conventional and organic avocado’s fluctuate throughout the year. There is a first peak around week 30. Organic avocado’s seem to have two peak more later in the year. Conventional avocado prices seem to go up with a peak around week 44, after week 44 the price for organic and conventional avocado’s drop.
2017
Average prices for avocado’s throughout the year has changed a lot compared to 2015. Bigger price differences throughout the year, for organic and conventional avocado’s. Week 9 is rather interesting as conventional and organic avocado prices come really close and there is almost no price different between them. The price seem to build up for both from week 10 onwards. Organic seems to peak slightly earlier in the year around week 34, conventional avocado’s seem to peak around week 40.
2018
The first 12 weeks for conventional avocado’s started off more or less the same as 2017. The price for organic started higher from the beginning of the year and look more stable in the first 12 weeks compared to 2017.
Conclusion – part 1
Conclusion so far is that throughout the year organic avocado’s are always more expensive although price differences might not be the same all the time any more. The price of an avocado fluctuates more. In their ‘down’ time, they become slightly cheaper, in their peak time, they become more expensive, where the organic avocado is on average being sold for over $2.00 in week 32-34 in 2017. Avocado’s seem to become more seasonal throughout the year. Stronger peaks towards the end of the year from week 30-35 onwards, lower prices at the end of the year and the beginning. It feels like prices seem to go up every year. But is that the case?
Seeing those changes in a chart makes me curious how strong the relation between those factors is, and whether the differences are significant or not. Therefore I’ll do some correlation tests between the avocado price and the type of avocado but also between years.
As we can see in de regression results (I marked the numbers that needs to be interpreted), both year (P>t = 0.000 of 2016 & 2017, P>t=0.021 on 2018) and type (P>t = 0.000) have a significant impact on average price per avocado. An organic avocado is significant more expensive (check coef = +0.496). On average $0.50 per avocado to be precise. The price of an avocado differs significantly every year. They are $0.04 cheaper in 2016 and $0.03 in 2018 (2018 is not complete so this doesn’t give you a full realistic overview). 2017 had the biggest increase with $0.17 more expensive avocado’s over all. As we have seen before, throughout the years there are big fluctuations in price difference. Where it seems to be the case that the average price of avocado’s was cheaper at some times, but also more expensive around other times of the year. This shows that only a regression or only a few charts does not give you enough information to know exactly what’s going on. Chart and regressions in this case give you a more complete overview and help interpret the data correctly.
R-square indicated for how much one variable explains for the result of the other. So, type of avocado explains for 37,9% the price of the avocado. Year explains for 4% the price of the avocado. Which makes sense based on the explanation previous about prices going down in 2016 and 2018, but up, in 2017. What if we combine both in one big regression? Will those two variables together explain the average price of an avocado?
Results type of avocado vs. average price per avocado
Results year vs. average price per avocado
Multivariate regression analysis
Below I mad a multi regression analysis. As you can see, all of the variables (Year2016, 2017, 2018 and Type (organic) have a significant impact (check P>[t]). All together they explain for 41,4% the average price of an avocado. Which is a higher percentage than just the type of avocado. So, year is an extra variable that needs to be taken into account.
After this analysis you could continue doing a regression analysis for all of the regions to see if there is one or a group of regions that stand out that explain the price of an avocado. For now I leave it for what it is. Seasonality is also an interesting conclusion of the charts. It would be very interesting to find out what impacts the seasonality. Unfortunately this data set does not contain enough information to explore and analyse that data.
Over all conclusion
Knowing this, I personally probably keep an eye out in the period roughly between week 35, 40. Just knowing that avocado’s are more expensive in that time of the year might be worth it considering leaving them for a few weeks, knowing that the price will drop. Of course, you don’t have to. If you just like avocado’s that much that you don’t really care, that’s fine too. But know at least you know.
Cheers,
Laura