Does the nutrient composition of vegetables reflect onto the price?
People pay more attention to their health these days. With that health-consciousness, there began to come new vegetable breeds enriching the specific nutrients with higher price in stores.
Generally, the price is to go up by demand increase with expectation for goods.We investigated whether vegetable prices are impacted with the expectation for nutritional value.
To be more specific, we have analyzed either the nutritional value of vegetables reflects onto the price besed on historical price data in vegetable market and ingredient data of vegetables.
We took historical Vegetables Market price data published by Ministry of Agriculture, Forestry and Fisheries(daily data).
Reference : https://www.seisen.maff.go.jp/seisen/bs04b040md001/BS04B040UC020SC998-Evt001.do
We have studied the daily data from January 2012 to October 2016, because there are less data published on the web page before 2011.
Ingredient data of vegetables are taken from 2015 year edition of Standard tables of food composition publised by Ministry of Education, Culture, Sports, Science and Technology.
Reference : http://www.mext.go.jp/a_menu/syokuhinseibun/1365297.htm
We made an interpolation with 0.00 for the ingredient which isn’t measured or the ingredient which doesn’t reach the minimum mentioned.
We ran our analysis for 74 items (48 vegetables and 26 fruits), which are availabe for both price data and ingredient data.
The nutrient information data is categorized according to the part and the recipe, then we took the data from main eatable parts and raw ingredient content. (for example, in case of Japanese radish, record of ” (Japanese radish kinds) Japanese radish root with peel and raw”).
When confirming the influence on price by the nutrient composition, it is necessary to use the price from which any factor extraneous to nutritional ingredients needs to be excluded.Because those prices moves day by day with the various factors which aren’t related to the ingredient content such as demand-supply and seasonality, etc.. We have applied the following modified model based on Nanseki’s market price estimation model  to run factor analysis on prices.
log(Pt) = a0 + a1log(Qt) + ΣdiDi + ΣciMi + ΣbiYi
Pt=vegetable price at t
Qt=Supply amount at t
Yi=dummy coefficient to denote year
Mi=dummy coefficient to denote month
Di=dummy coefficient to denote day of the week
We could say a0 is expressing facor which has nothing to do with demand-supply or seasonality.
We have run this price factor analysis for each vegetables and get a0 for each.Furthermore, we have run correlatoin analysis between a0 and ingredient data,item by item, to see either there is any relation between them.
You will find the precision of our market price estimation model ,item by item,in Table1.
There is dummy coefficient to denote month (Mi) to express seasonality in this model,the coefficient of determination for some items are not so high.This is because Mi is computed as monthly mean even for prices of some items whose season factor drastically change in the same month.
We have made the futher analysis for the all items regardless of the coefficient of determination. It is not necessary to consider how well our model explains the price movement because our main purpose is to take out the impact which isn’t related to the ingredients and not to estimate daily price behavior.
However we have not run further analysis for the items whose calcuated parameter a1 is positive which means supply increase causes price up.This is not reasonalble result so we think either price estimation model or historical data is not appropriate.
You will find the correlation analysis between a0 and the content of each nutrient for 49 items in Table2.
We have tried correlation absense test for all to find only β-tocopherol shows significant correlation between them with p<0.05. However 2 items,”Japanese Yam” and “loquat” have β-tocopherol, so it may be resulted due to outlier, it can’t be said that the content has an influence on the price.
We can think there is no influence the ingredient content exerts on the price.
Previous contents : Price Mechanism for Vegetables in Japan (Part 2)
 Teruaki Nanseki(2000)「日別野菜価格変動の長期的変化と要因分析」『農業経営研究』第 38 巻 1号 pp.1-10
 Akio Sakura（1990）「野菜卸売市場における旬別価格の時系列変動要因分析 一番黄浜市中央卸売市場本場を中心として」『神奈川県農業総合研究所研究報告』第132号 pp.62-68
Mitsuo Suzuki(2004)「東京都中央卸売市場における野菜卸売価格変動の要因分析 ―キャベツと キュウリを対象として―」『農村研究』第 98 号 pp.34-44