We investigate the power of time series analysis based on a variety of information-theoretic approaches from statistics (AIC, BIC) and machine learning (Minimum Message Length) - and we then compare their efficacy with traditional time series model and with hybrids involving deep learning. More specifically, we develop AIC, BIC and Minimum Message Length (MML) ARMA (autoregressive moving average) time series models - with this Bayesian information-theoretic MML ARMA modelling already being new work. We then study deep learning based algorithms in time series forecasting, using Long Short Term Memory (LSTM), and we then combine this with the ARMA modelling to produce a hybrid ARMA-LSTM prediction. Part of the purpose of the use of LSTM is to seek capture any hidden information in the residuals left from the traditional ARMA model. We show that MML not only outperforms earlier statistical approaches to ARMA modelling, but we further show that the hybrid MML ARMA-LSTM models outperform both ARMA models and LSTM models.

Repeated fires cause peatlands to degrade. Fire management has been carried out, but fires continue to occur, especially during the dry season. Through social capital that applies in the community based on the local wisdom of the Banjar people in environmental management, it is hoped that peatland fires can be overcome in this area. This research aims to analyze the social capital of the Banjar people in mitigating land fires based on local wisdom and the environment. The sample in this study was 250 people who live in an area prone to fires on peatlands in Banjarbaru, South Kalimantan, Indonesia. There are three variables in this study, namely Trust (T), Network (N) and Norm (N). There are 33 Likert scale questions (1-4, strongly agree-strongly disagree). We also conducted interviews with 20 members of the Fire Care Community (MPA) to obtain information about the social capital of the Banjar community in mitigating land fires. The analysis used is Structural Equation Modeling (SEM). The study results show norms have the greatest influence on disaster mitigation of peatland fires compared to trust and social networks. Norms in the community in the form of values from peatlands, rules, norms and sanctions regulate the community more to deal with the problem of peatland fires than trust and social networks