Page 443 - Contributed Paper Session (CPS) - Volume 4
P. 443
CPS2564 Tiffany Rizkika et al.
Nowcasting modelling of volatile and non-
volatile food prices using crowdsourcing data
(case study of some food commodities prices on
Lombok island in 2015)
1
2
3
3
Tiffany Rizkika , Wida Siddhikara P , Budi Yuniarto , Ricky Yordani , Setia
Pramana
3
1 BPS Solok Selatan
2 BPS Statistics Indonesia
3 Politeknik Statistika STIS
Abstract
Monitoring of food prices in real-time or called nowcasting is important to
maintain price transparency so that the government easy to detect inflation
early and the economy becomes more stable and also can reduce the potential
of economic turmoil. This research investigate the usage of crowdsourcing
data to nowcast the volatile and non-volatile food prices in Mataram,
Indonesia. This research uses two approaches to nowcast the prices, i.e.,
historical data-based to nowcast volatile food prices, and present data-based
to nowcast non-volatile food prices. Historical data-based uses two methods,
namely statistical modelling using Distributed Lag Model and machine
learning using Neural Network RPROP, also using two types of periods daily
and weekly. The result of this approaches shows that the suggested model is
Distributed Lag Model, and the recommended data period is weekly. While
nowcast based on present data uses two methods, they are time series-based
(Nowcast Model) and statistical filtering-based which is followed by cubic
smoothing spline modelling (IQR-Spline Model, KDE-Spline Model). We found
that the IQR-Spline model is better than KDE-Spline for the commodities of
long beans, and mackerel. The KDE-Spline model is better than IQR-Spline for
the commodities of instant dry noodles, peanuts, and vegetable tomatoes. In
time series-based models, the nowcast model with parameter modification
has better results than the nowcast model with parameters as in the previous
studies.
Keywords
Nowcasting; Food Price; Volatile; Non-Volatile
1. Introduction
Transparency of food prices is needed by government to making a right
decision about food price policy. The right decision of this essential parts of
economy, can reduce the potential of economic turmoil. To monitor the price
of goods and services at consumer level, BPS-Statistics of Indonesia conducts
Consumer Price Survey collecting consumer price data from markets. The
432 | I S I W S C 2 0 1 9