The purpose of the study is to forecast the price of rice in the city of Denpasar in 2017 using backpropagation neural network method. Backpropagation neural network is a model of artificial neural network by finding the optimal weight value. Artificial neural networks are information processing systems that have certain performance characteristics similar to that of human neural networks. This analysis uses time series data of rice prices in the city of Denpasar from January 2001 until December 2016. The results of this research, concludes that the lowest rice price is predicted in July 2017 at Rp9791.5 while the highest rice price in April 2017 for Rp9839.4.

Singular spectrum analysis (SSA) is a method to decompose the original time series into a summation of a small number of components that can be interpreted as varied trends, oscillatory, and noise components. The purpose of this research is to model and to find out the results of forecasting the number of foreign tourists arrival to Bali using SSA method. In this research, the accuracy of forecasting results will be calculated using the SSA model with reccurent singular spectrum analysis (RSSA) method. The best SSA model was obtained with a window length (L=94) and produces MAPE value of 7,65%.

Copula is already widely used in financial assets, especially in risk management. It is due to the ability of copula, to capture the nonlinear dependence structure on multivariate assets. In addition, using copula function doesn't require the assumption of normal distribution. There fore it is suitable to be applied to financial data. To manage a risk the necessary measurement tools can help mitigate the risks. One measure that can be used to measure risk is Value at Risk (VaR). Although VaR is very popular, it has several weaknesses. To overcome the weakness in VaR, an alternative risk measure called CVaR can be used. The porpose of this study is to estimate CVaR using Gaussian copula. The data we used are the closing price of Facebook and Twitter stocks. The results from the calculation using 90% confidence level showed that the risk that may be experienced is at 4,7%, for 95% confidence level it is at 6,1%, and for 99% confidence level it is at 10,6%.

Poisson regression is a nonlinear regression that is often used to model count response variable and categorical, interval, or count regressor. This regression assumes equidispersion, i.e., the variance equals the mean. However, in practice, this assumption is often violated. One of this violation is overdispersion in which the variance is greater than the mean. There are several  methods to overcome overdispersion. Two of these methods are negative binomial regression and generalized Poisson regression. In this research, binomial negative regression and generalized Poisson regression statistically equally good in handling overdispersion.

The purpose of this research is to compare the selling price of down and out barrier option when the prices are simulated by the Antithetic Variate Monte Carlo and the standar Monte Carlo. Barrier options are path dependent options and the payoff depend on whether the underlying asset price touched the barrier or not during the life of the option. In this research, we conducted simulations against the closing price of the shares of PT Adhi Karya using Standard Monte Carlo simulation and the Monte Carlo-Antithetic Variate simulation. After the simulation, we obtained that the option prices using Antithetic Variate produces a cheaper price than the standar one. We also found that the analytic solution has a smaller error on its confidence interval compare to the Monte Carlo Standar.