Consumer price index prediction using Long Short Term Memory (LSTM) based cloud computing

Abstract

Long Short Term Memory (LSTM) is known as optimized Recurrent Neural Network (RNN) architectures that overcome the lack of RNN’s about maintaining long period of memories information. As part of machine learning networks, LSTM also notable as the right choice for time-series prediction. Inflation rate has been used for decision making for central banks also private sector. In Indonesia, CPI (Consumer Price Index) is one of best practice inflation indicators besides Wholesale Price Index and The Gross Domestic Product (GDP). Since CPI data could be used as a direction for next inflation move, we conducted CPI prediction model using Long Short Term Memory Method. The network model input consists of 34 variables of staple price in Surabaya and the output is CPI value. In the interest of predictive accuracy improvement, we used several optimization algorithm i.e. Stochastic Gradient Descent (sgd), Root Mean Square Propagation (RMSProp), Adaptive Gradient (AdaGrad), Adaptive moment (Adam), Adadelta, Nesterov Adam (Nadam) and Adamax. The result indicate that Nesterov Adam has 4.088 RMSE’s value, less than other algorithm which indicate the most accurate optimization algorithm to predict CPI value.

Forecasting The Consumer Price Index: A Comparative Study of Machine Learning Methods

“…More recently, researchers who used machine learning methods include Huong et [21] who forecasted the CPIs of AU, Spain and OECD countries (550 data values) with an ensemble learning model and the NSGA-II multi-objective evolutionary algorithm, and only reported MSE numbers to evaluate the goodness of their forecast. Zahara et al [22] used the long short-term memory (LTSM) deep learning technique to predict Indonesian CPI. For optimization, they used stochastic gradient descent (SGD), and a few other methods and concluded that the Adaptive moment (Adam) optimization algorithm resulted in the best RMSE.…”

Food price index prediction using time series models: A study of Cereals, Millets and Pulses

“…First step is that, the time series is decomposed into its three components -error, trend, and seasonality. This is done using a method such as seasonal decomposition of time series (STL) [7]. Once the components are identi ed, the next step is to model each component separately.…”

China’s GDP forecasting using Long Short Term Memory Recurrent Neural Network and Hidden Markov Model

“…Gross Domestic Product (GDP) is a key indicator of economic growth, which measures the total value of goods and services produced within a country in a year, not including its income from investments in other countries. There are considerable literatures on economic prediction in financial market that utilises Hidden Markov Model (HMM) [1,2] and Long Short-Term Memory (LSTM) recurrent neural network [3,4]. However, there isn't any study on the application of LSTM-HMM in GDP forecast that combines the potentials of these two models, which motivates our research on the China's GDP forecast using LSTM-HMM considering the relationship between inflation and economic growth.…”

“…Long Short-Term Memory (LSTM) is a type of recurrent neural network that works better on tasks involving long time lags by bridging huge time lags between relevant input events [14], which has emerged as an effective and scalable model for time-series prediction. LSTM is applied to CPI prediction in Indonesia with multivariate input [3]. LSTM is also utilised to the optimal hedging in the presence of market frictions [4], which shows usefulness in the empirical analysis of real option markets.…”