In this study, it has attempted both univariate and multivariate approaches of time series analysis. This study has conducted at Slyhet division in Bangladesh. Firstly, we discussed the methodologies that are used in this study. Secondly, we have shown the univariate time series (monthly maximum temperature, monthly minimum temperature, monthly maximum humidity, monthly minimum humidity, monthly average cloud amount, monthly average rainfall, monthly average sea level pressure in Sylhet) graphically to study their descriptive statistics. Thirdly, we have fitted SARIMA model for each of the univariate time series. The fitted models for monthly maximum temperature, monthly minimum temperature, monthly maximum humidity, monthly minimum humidity, monthly average cloud amount, monthly average rainfall and monthly average sea level pressure are ARIMA (1,0,0)?(2,0,0)12 , ARIMA (2,0,0)?(2,0,0)12 , ARIMA (3,1,0)?(2,0,0)12 , ARIMA (0,0,0)?(2,0,0)12, ARIMA (0,0,0)?(2,0,0)12, ARIMA (1,0,0)?(2,0,0)12 and (1,0,0)?(2,0,0)12 respectively. We have constructed the VAR (2) model which includes seven equations for the seven variables. These all models have successfully passed the diagnostic stage.
Climate change will impact horticultural production in the future. The present work assesses the future climatic impact on regional horticultural production by establishing a basic frame of a climate impact modeling chain. Using high resolved simulated climate time series of future alternatives of the worlds development, long-term trends of various climate effects were assessed. For this purpose, simulated climate time series were calibrated and effects of resolution, bias, bias correction, scenario, climate model and impact model as well uncertainty propagation along the simulation chain were investigated. A multidimensional bias correction method was developed in order to optimize climate time series consistency.
Detection and attribution of climate change dependexplicitly on proper characterization of naturalpatterns of climate over time. This book examinesmultiple types of climate data, from ocean heatcontent over a few years to ice core data going back60,000 years. Multi-scale analyses are employed totest for stationarity or acceleration of warming insatellite and ground instrumental data. Problemsthat arise when combining instrumental data into longrecords are evaluated. The CO2 rise pattern ischaracterized and the problem of extrapolation isaddressed. It is finally shown that periodicpatterns can be detected at multiple time scales,from decades to millenia. These patterns arecharacterized and used to address the attributionproblem. The human contribution to climate is shownto be separable from natural variability and ischaracterized. Human forcing of climate is shown tobe detectable beginning in 1942. This forcingpattern is used along with patterns of naturalvariability to make a 100 year forecast.
Increasing temperatures in the Lawra District of the Upper West Region (the driest region in Ghana) has caused devastating flash floods resulting from high intensity short duration rainfalls. These are characterized by annual mean rainfall intensities above 1000mm. However using Digital Elevation Model (DEM) for Hydrological Modelling and watershed delineation with remote sensing data and GIS techniques, for flood forecasting, requires an understanding of the basic relationships between rainfall and runoff for effective management of flood water. Eight Statistical tools; Time Series, Trend Analysis, Moving Average, Weighted Moving Average, Exponential Smoothing, Percentage Growth and Seasonality, are statistical analyses conducted on 31 years of data each on temperature, rainfall and flow rates for their correlation and variability, in predicting these floods in an ever changing climate scenarios on the Black Volta.
Time series models can be used for generating data for planning and design or estimating the future outcomes of the processes. For the stochastic modeling of hydrologic time series, various models are offered depending on the type of the time series (yearly, monthly, daily, etc.). In stochastic modeling of time series in hydrology, one of the encounter important problem encountered is the choice of the best model type between alternative models.
The study examines long-term urban modification of mean annual conditions of surface temperature and rainfall in the city of Maseru, Lesotho. This is done by studying trends in the minimum and maximum temperatures, rainfall and number of rain days time series. Data from an urban station in Maseru and three rural stations was used. The data comprised minimum and maximum temperatures for the period 1970-2005 and daily rainfall figures for the period 1922-2005. Results showed that abrupt climatic change in temperatures towards warming occurred at the urban station than at the rural station. This was observed to be larger in the minimum temperature series and this began in the early 1990s. No significant changes in rainfall for the period 1922-2005 were found, except for short time periods within the series. The warming trends in minimum temperature, pose potential challenges on climate and urban planning of the city. In particular the effect of increased minimum temperature will influence human physiological comfort, building and urban design, wind circulation and air pollution. The findings in this study can be used to inform future planning of the city.
This book presents a new model of neural network for time series analysis and forecasting: the ARMA-CIGMN (Autoregressive Moving Average Classical Incremental Gaussian Mixture Network) model and its analysis. This model is based on modifications made to a reformulated IGMN, the Classical IGMN (CIGMN). The CIGMN is similar to the original IGMN, but based on a classical statistical approach. The modifications to the IGMN algorithm were made to better fit it to time series. The ARMA-CIGMN model demonstrates good forecasts and the modeling procedure can also be aided by known statistical tools as the autocorrelation (acf) and partial autocorrelation functions (pacf), already used in classical statistical time series modeling and also with the original IGMN algorithm models. The ARMA-CIGMN model was evaluated using known series and simulated data.
This thesis has considered RCM (Regional Climate Data) which is 50km by 50km grid resolution and more accurate to change or to downscale the grid data in to point climate data(station data)than the Global model(GCM) which has the same grid resolution. This thesis work is possible by the help of almighty God and my family specially my sisters Kassaye, Bezu and Nigist Belay and my advisor Dr. Semu Ayalew.
This research investigates the economic impact of climate variability on Banana production in the Cameroon Development Corporation (CDC). The study used a time series analyses to estimate the effects of rainfall, temperature and wind speed variability on banana production. The research also used the Lorenz curve to analyse climate variability over time and a spatial analogue methodology to estimate the effects of climate variability on Banana production in the CDC. Using two structural equations on a 37-years data set, the findings of this research showed that rainfall, temperature and wind speed variability have a significant effect on banana production. It also revealed that climate variability is relatively more significant in determining banana output than capital and labour. A test for structural defects in the model revealed also that there is a sharp increase in rainfall temperature and wind speed variability between 1985 and 2009. As a result, the study recommended irrigation practices to mitigate the adverse effects of the rainfall, temperature and wind speed variability.
This monograph provides Bayesian inference for change point problems through Mixture model approach in Time series models viz., changes in mean of the time series with and without auto correlated errors, variance changes in the time series model and order changes in the time series models. MCMC technique is used to obtain the numerical solutions. The main aim of the numerical study is to illustrate the evaluation of the estimates of the parameters on the basis of the methodology developed in this monograph.
This book attempts to develope some new inferential procedures for time series regression models.An inferential method for a time series linear regression model with auto correlated disturbances using quarterly data, has been developed by proposing a test based on internally studentized residuals.Two modified estimation procedures have been proposed for time series regression models involving MA (1) and MA (q) process errors.Autoregressive moving averages and autoregressive conditionally heteroscadastic (ARCH) processesses have been specified systematically with their characteristics. The generalized ARCH model is specified and the effect of error structure on ARCH model has been explained. Two modified tests for detecting the problem of ARCH errors have been developed by using Box-pierce-lying test statistics based on internally studentized residuals. A new estimation procedure has been developed for ARCH model by using an interactive technique
Information on climate and its variability during the different time scale is important to understand the various internal and external forcing on the climate and thereby make a reliable future prediction. The present knowledge of climate variability, climate change is based upon the observed meteorological data. The information beyond the instrumental period is also very important to understand the long term climate change variability on decadal to century scale. However, such past climate data information is limited and discontinuous. Very little information is available on Indian climate, particularly over the Himalayan region earlier to about a century when the instrumental records commenced. The tree-ring records, though shorter in palaeoclimatic time frame, are accurate and their time resolution is to a specific season or year. Preliminary studies based on tree-ring samples in the region beyond 500 years in age with well marked and environmentally sensitive annual growth rings and can be used to reconstruct past climatic variation over the region.
Time series are a special form of data where past values in the series may influence future values, depending on the presence of underlying deterministic forces. These forces may be characterised by trends, cycles and nonstationary behaviour in the time series and predictive models attempt to recognise the recurring patterns and more particularly potential linear or nonlinear relationships between past and actual values, or with other exogenous variables which may be linked to the variable studied. Time series forecasting is the use of a model to forecast future time series values based on known past events: to predict data points before they are measured. Forecasting is an important and recurrent issue in business world since good forecasting models can lead to a major position in the market. Indeed a firm can anticipate the temporal evolution of a given data in order to implement solutions before its competitors. Forecasting problems find their applications in many fields: for example sales in marketing, production volume in operations and logistics, economic variable like GDP in macroeconomic studies or financial variables like stock prices in finance.
To assess how stream flow in Gilgel Abbay River Basin will be affected by climate change, the HadCM3 model, developed at the Hadley Centre in the United Kingdom, was used to generate medium-high and medium-low emission scenarios in this study. The statistical downscaling model was used to generate future possible local meteorological variables in the study area. The down-scaled data were then used as input to the Soil and Water Assessment Tool hydrological model to simulate the corresponding future stream flow regime in Gilgel Abbay River Basin. Three benchmark periods simulated were 2011–2040 (2020s), 2041–2070 (2050s), and 2071–2099 (2080s). The time series generated by HadCM3 and statistical downscaling method indicate a significant increasing trend in both maximum and minimum temperature values, and a decreasing trend in precipitation. The hydrologic impact analysis made with the downscaled temperature and precipitation time series as input to the SWAT model suggested an overall decreasing trend in annual and monthly stream flow in the study area, in three benchmark periods in the future. This should be considered by policymakers of water resources planning and management.
Agriculture is always vulnerable to unfavorable weather events and climate conditions. The impacts of climate change on agriculture crop production are global concerns as well as Bangladesh. To measure Climatic and Hydrological effects on different types of crop productions in Bangladesh, Multiple Regression Model has been used as a measuring tools of cause-effect relation. At the same time, Auto Regressive Integrated Moving Average with external regressor, that is, ARIMAX model has used considering the time effects, because the data-set used in this study is a time sequence data. This is completely a new study for measuring the Climatic and Hydrological effects by using ARIMAX model. Again, from the comparative study, Multiple Regression is the best model for vegetable, potato and cereal production; and ARIMAX is the best model for rice, jute and species production for measuring the climatic and hydrological effects on agricultural production in Bangladesh.