Free Essay

Time Series

In:

Submitted By dock
Words 2184
Pages 9
I. Abstract
The United States and many countries around the world have been researching and developing methods of producing and utilizing alternative, cleaner fuel sources for many years. Natural gas is one of which that is being explored and used more and more on a daily basis. According to the American Petroleum Institute, Natural Gas meets 24% of U.S. energy demand and heats 51% of U.S. households. It is also a proven cleaner fuel source for automobiles when compared to gasoline or diesel. Many enterprises now employ the use of natural gas-powered vehicles for their demand, and there are now around 120,000 on U.S. roads (American Petroleum Institute).

With the ever-increasing presence of natural gas in our lives, it is important to know how it affects us on the consumer level. More importantly, it is essential to understand how it affects us monetarily. When we can understand trends and forecast the pricing of natural gas, we can improve our financial planning, resource allocation, etc.

II. Data Set Description and Methods Used

This paper will be conducting a time series analysis on U.S. Natural Gas prices from January 2002 through December 2012. Prices were collected by the Energy Information Administration on a monthly basis, and the prices are measured in dollars per thousand cubic feet. The raw data set is much more extensive, and measures data such as wellhead price, import and export price, etc. I chose to neglect this in my data set for simplicity sake and because it is more applicable to look at prices directly affecting everyday individual consumers.

The methods used in the time series analysis of U.S. Natural Gas Prices include non-seasonal differencing, exploring autocorrelation and partial autocorrelation functions, ARIMA modeling, conducting diagnostics on modeling by looking at residual normality [QQ plot, Shapiro Test], Box-Pierce test for autocorrelation, and other methods as well.

III. Detailed Results

(Please see below)

This is a time series plot of U.S. Natural Gas Prices from January 2002 – December 2012. As we can see, there is an apparent trend and seasonality present that we need to identify and eliminate before we develop an ARIMA model and forecast. Next, we will look at the ACF and PACF of the original data:

Looking at the ACF, we can see that the data is non-stationary because of the slow decay and sinusoidal pattern. It also confirms that seasonality is present. The PACF shows a significant lag at 1, so we will difference the data by lag of 1 and re-examine the differenced data.

The two displays below show the time series plot of the differenced data, and we can see that the trend is removed and hovering around 0. The ACF and PACF show that the once-differenced data is now stationary, with the ACF sinusoidal pattern around 0 and sharply declining.

Now we can start exploring an ARIMA model. We will perform this on the original data and incorporate our differencing we just did. Looking at the non-seasonal portion of the data, ACF, and PACF, we can start with a (1,1,0) explanation. With early tapering in the ACF and a clear spike at lag 1 in the PACF, I decided to start with a (1,1,0) non-seasonal portion. Looking at the seasonal behavior, I decided to start with a (2,0,0) explanation because no seasonal differencing is required. The ACF spikes at 1, 2, 3, 4 show seasonality, and seasonal spikes in 1,2 in the PACF indicate a (2,0,0) is a good option.

When we apply the (1,1,0) (2,0,0) modeling within 12 periods (because we are looking at monthly data), we get the following:

Series: dat.ts
ARIMA(1,1,0)(2,0,0)[12]

Coefficients: ar1 sar1 sar2 0.3443 0.3783 0.4501
s.e. 0.0884 0.0743 0.0770

sigma^2 estimated as 0.354: log likelihood=-124.5
AIC=256.99 AICc=257.31 BIC=268.49

In-sample error measures: ME RMSE MAE MPE MAPE MASE
0.0002117388 0.5927610268 0.4544843178 0.0049059772 3.5929656151 0.5232221253

Out of the different models tried, this one had one of the lowest AICc. The coefficients also confirm that this may be a valid model. However, we need to confirm that this is a valid model.

Shown below are some diagnostics to check. The first is standardized residuals over time. Outside of a few outliers, this shows constant variance. The ACF of residuals show no significant lags (there are some ones slightly larger than we like). We will examine this further below:

The ACF and PACF confirm that we may continue with our model diagnostics. There may be some white noise shown here, but in the grand scheme of things it is very minimal in terms of affecting our overall model. We now want to see that the residuals are normal… this is done through the QQ plot shown below:

The QQ plot may show that the residuals of our modeled data are normal, but this is not enough. We need to perform a Sharpiro-Wilk normality test to confirm this: Shapiro-Wilk normality test data: residuals(mod.1) W = 0.9836, p-value = 0.1143 The test confirms normality as we fail to reject the null hypothesis and conclude normality of residuals. Our next diagnostic is confirming stationarity through the Augmented Dickey-Fuller Test: Augmented Dickey-Fuller Test data: residuals(mod.1) Dickey-Fuller = -5.8323, Lag order = 5, p-value = 0.01 alternative hypothesis: stationary The ADF test shows stationarity. We now want to perform a Box-Pierce on our modeled data to see if there is any autocorrelation within the residuals: Box-Pierce test data: residuals(mod.1) X-squared = 8e-04, df = 1, p-value = 0.9776 This high box-pierce p-value is desirable, as it indicates no autocorrelation. Now that we have our model and confirmed it is valid, we can now forecast U.S. Natural Gas prices for the next two years (h=24) and show the prediction intervals: Point Forecast Lo 80 Hi 80 Lo 95 Hi 95 Jan 2013 9.205785 8.501290 9.91028 8.128353 10.28322 Feb 2013 9.194351 8.278040 10.11066 7.792974 10.59573 Mar 2013 9.323437 8.219602 10.42727 7.635268 11.01161 Apr 2013 9.873939 8.545128 11.20275 7.841699 11.90618 May 2013 10.706935 9.111116 12.30275 8.266340 13.14753 Jun 2013 11.908033 9.975870 13.84020 8.953044 14.86302 Jul 2013 12.778663 10.548804 15.00852 9.368388 16.18894 Aug 2013 13.029919 10.606853 15.45299 9.324159 16.73568 Sep 2013 12.723313 10.219584 15.22704 8.894190 16.55244 Oct 2013 11.182895 8.867335 13.49846 7.641551 14.72424 Nov 2013 9.957113 7.797628 12.11660 6.654465 13.25976 Dec 2013 9.364286 7.245231 11.48334 6.123471 12.60510 Jan 2014 9.230339 7.057970 11.40271 5.907988 12.55269 Feb 2014 9.218187 6.968090 11.46828 5.776960 12.65941 Mar 2014 9.346931 6.986332 11.70753 5.736707 12.95715 Apr 2014 9.898123 7.317101 12.47914 5.950791 13.84546 May 2014 10.732425 7.848270 13.61658 6.321490 15.14336 Jun 2014 11.935590 8.635448 15.23573 6.888458 16.98272 Jul 2014 12.807407 9.169267 16.44555 7.243352 18.37146 Aug 2014 13.058408 9.252454 16.86436 7.237704 18.87911 Sep 2014 12.750354 8.942053 16.55865 6.926061 18.57465 Oct 2014 11.205997 7.779714 14.63228 5.965949 16.44604 Nov 2014 9.977107 6.857425 13.09679 5.205965 14.74825 Dec 2014 9.382564 6.385021 12.38011 4.798217 13.96691

R-Code Input: dat = read.table("clipboard", header = TRUE) head(dat); tail(dat); names(dat) dat.ts = ts(dat$Price, frequency = 12, start=c(2002,1)) plot(dat.ts) points(dat.ts) par(mfrow=c(2,1)) acf(dat.ts, lag.max = 50) pacf(dat.ts, lag.max = 50) par(mfrow=c(1,1)) d.ts.diff = diff(dat.ts, lag=1, differences=1) plot(d.ts.diff) par(mfrow=c(2,1)) acf(d.ts.diff, lag.max = 50) pacf(d.ts.diff, lag.max = 50) par(mfrow=c(1,1)) mod.1 = arima(dat.ts, order = c(1,1,0), seasonal = list(order = c(2,0,0), period = 12)) summary(mod.1) tsdiag(mod.1) par(mfrow=c(2,1)) acf(residuals(mod.1)) pacf(residuals(mod.1)) par(mfrow=c(1,1)) library(car) library(quadprog) library(tseries) library(forecast) qqPlot(residuals(mod.1)) shapiro.test(residuals(mod.1)) mod.1.fit = fitted(mod.1) adf.test(residuals(mod.1)) Box.test(residuals(mod.1)) forecast(mod.1.fit, h = 24) plot(forecast(mod.1.fit, h = 24)) Data Used: Month Price Jan-2002 7.38 Feb-2002 7.23 Mar-2002 7.1 Apr-2002 7.66 May-2002 8.54 Jun-2002 9.58 Jul-2002 10.31 Aug-2002 10.44 Sep-2002 10.23 Oct-2002 8.61 Nov-2002 7.99 Dec-2002 7.87 Jan-2003 8.18 Feb-2003 8.58 Mar-2003 9.77 Apr-2003 10.18 May-2003 10.79 Jun-2003 12.08 Jul-2003 12.75 Aug-2003 12.84 Sep-2003 12.31 Oct-2003 10.64 Nov-2003 9.77 Dec-2003 9.51 Jan-2004 9.71 Feb-2004 9.85 Mar-2004 10.03 Apr-2004 10.54 May-2004 11.63 Jun-2004 13.08 Jul-2004 13.54 Aug-2004 13.74 Sep-2004 13.31 Oct-2004 11.69 Nov-2004 11.44 Dec-2004 11.09 Jan-2005 10.9 Feb-2005 10.87 Mar-2005 10.84 Apr-2005 11.88 May-2005 12.74 Jun-2005 13.79 Jul-2005 14.86 Aug-2005 15.51 Sep-2005 16.56 Oct-2005 16.44 Nov-2005 15.64 Dec-2005 14.6 Jan-2006 14.92 Feb-2006 13.98 Mar-2006 13.17 Apr-2006 13.27 May-2006 14.41 Jun-2006 15.07 Jul-2006 15.72 Aug-2006 16.18 Sep-2006 15.71 Oct-2006 12.51 Nov-2006 12.45 Dec-2006 12.53 Jan-2007 12.17 Feb-2007 12.13 Mar-2007 12.81 Apr-2007 13.31 May-2007 14.69 Jun-2007 16.28 Jul-2007 16.71 Aug-2007 16.71 Sep-2007 16.03 Oct-2007 14.57 Nov-2007 13.04 Dec-2007 12.34 Jan-2008 12.24 Feb-2008 12.58 Mar-2008 13.13 Apr-2008 14.49 May-2008 16.33 Jun-2008 18.91 Jul-2008 20.77 Aug-2008 20.17 Sep-2008 18.41 Oct-2008 15.45 Nov-2008 13.8 Dec-2008 12.84 Jan-2009 12.49 Feb-2009 12.26 Mar-2009 11.98 Apr-2009 11.68 May-2009 12.86 Jun-2009 14.26 Jul-2009 15.27 Aug-2009 15.61 Sep-2009 14.8 Oct-2009 11.78 Nov-2009 11.48 Dec-2009 10.42 Jan-2010 10.56 Feb-2010 10.69 Mar-2010 10.99 Apr-2010 11.97 May-2010 13.12 Jun-2010 14.86 Jul-2010 16.21 Aug-2010 16.65 Sep-2010 15.63 Oct-2010 13.37 Nov-2010 10.89 Dec-2010 9.98 Jan-2011 9.9 Feb-2011 10.14 Mar-2011 10.43 Apr-2011 11.27 May-2011 12.5 Jun-2011 14.7 Jul-2011 16.14 Aug-2011 16.67 Sep-2011 15.63 Oct-2011 12.85 Nov-2011 10.78 Dec-2011 9.83 Jan-2012 9.67 Feb-2012 9.52 Mar-2012 10.45 Apr-2012 11.01 May-2012 12.66 Jun-2012 14.25 Jul-2012 15.2 Aug-2012 15.89 Sep-2012 14.81 Oct-2012 11.78 Nov-2012 10.06 Dec-2012 9.75 Sources: http://www.api.org/oil-and-natural-gas-overview/exploration-and-production/natural-gas/natural-gas-uses http://www.eia.gov/energyexplained/index.cfm?page=natural_gas_factors_affecting_prices RAW DATA SET OBTAINED FROM : http://www.eia.gov/naturalgas/data.cfm#prices

R SESSION INFO: R version 2.13.0 (2011-04-13) Platform: i386-pc-mingw32/i386 (32-bit) locale: [1] LC_COLLATE=English_United States.1252 [2] LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 [4] LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] splines stats graphics grDevices utils datasets methods [8] base other attached packages: [1] forecast_3.14 RcppArmadillo_0.2.34 Rcpp_0.9.10 [4] fracdiff_1.4-0 tseries_0.10-28 zoo_1.7-7 [7] quadprog_1.5-4 car_2.0-11 survival_2.36-5 [10] nnet_7.3-1 MASS_7.3-12 loaded via a namespace (and not attached): [1] grid_2.13.0 lattice_0.19-23 tools_2.13.0

Similar Documents

Premium Essay

Time Series and Forcasting

...Time Series and Forecasting Learning Team A Quantitative Reasoning for Business/501 August 23, 2011 Dr. Champion Time Series and Forecasting The purpose of this paper is for statisticians from the accountant department of Norton Company to compute the quarterly seasonal index for the years of 2003 through 2006 by using the ratio-to-moving-average method. In addition, the accountants will deseasonalize the data and determine the trend equation. Furthermore, the statisticians will estimate Norton Company’s seasonally adjusted sale for the four quarters of 2007. The quarterly sales for the Norton Company were given in millions of dollars for four years. Therefore, the recorded quarterly sales for the Norton Company were referenced and included 16 quarters total. A scatter plot was graphed showing the recorded sales for the 16 quarters. The following equation was obtained from the historical data: y = 0.461x + 4.2 with the R² of 0.1695 and a R of 0.412. R is the coefficient of correlation, which provides the strength and direction of a linear relationship. R2 is the coefficient of determination, which measures the amount of variation in y that can be explained by the variation in x for example: 0 ≤ R2 ≤ 1. With this unadjusted regression equation and R² we can see that there is seasonality. Accordingly, a higher R² is required to prove that our equation is accurate and good for forecasting. The first step is to find the moving total....

Words: 840 - Pages: 4

Free Essay

Time Series Analysis

...Time series analysis We are pleased to submit the following report on the “Time Series Analysis”. By completing the report, we have got acquainted importance and relevance of time series on business application. We also perceived idea on the whole process of Time Series Analysis. We acquired knowledge about the method of measuring trend, growth rate, acceleration rate etc. In spite of limitation of time & opportunity we have tried our level best to complete the report. We are pleased to provide you with this report with necessary analysis, references and we shall be available for any clarification, if required. Thank you for assigning us in this study. On behalf of the group Md. Arif Hasan ID: 12-150 Table of Contents Serial No Topic Page No 1 Letter of Transmittal 1 2 Rationale of the study 2 3 Objectives of the report 3 4 Methodology of the report 3 5 What is Time Series 4 6 Uses of Time Series in Business 5 7 Components of a Time Series 5 8 Classical Time Series Model 9 9 Methods of trend measurement 9 10 Least squares method 10 11 The Growth Rate 14 12 The Acceleration Rate 15 13 Rule of 72 16 14 Bibliography 17 Rationale of the study Having been assigned to prepare a report on Time Series Analysis we are submitting the term paper based on our findings and understandings. Time series analysis has vast application and is of huge importance in the field of Business and Economics as well as in decision making thereof. Calculating secular trend we can...

Words: 2908 - Pages: 12

Free Essay

Time Series Analysis

...Time Series Analysis Yt=observed value of the time series in time period t TRt=the trend component or factorin time period t SNt=the seasonal componentor factorin time period t CLt=the cyclical componentor factorin time period t IRt=the irregular componentor factorin time period t 7.1) CL*IRCL=IR a) SN1=1.191 TR1=240.5 CL1=null IRt=null SN2=1.521 TR2=260.4 CL2=0.998 IR2=0.990 SN3=0.804 TR3=280.4 CL3=0.994 IR3=0.986 SN4=0.484 TR4=300.3 CL4=1.003 IR4=1.008 b) It presents a multiplicative decomposition model Yt=TRt*SNt*CLt*IRt SNt*IRt=YtTRt CLt snt*irt=YtCMAt =YtCMAt Equation of the estimated trend: TRt=Bo+B1t dt=B0+B1t+εt TRt=220.53+19.94(t) c) Yt=trt*snt Y17=220.53+19.9417*1.191=666.6 Y18=220.53+19.9418*1.521=881.6 Y19=220.53+19.9419*0.804=482.1 Y20=220.53+19.9420*0.484=299.9 d) Yt=trt*snt*cl We cannot see a definitive cycle and because the values of cl are close to 1. We do not take it into account. Y21=220.53+19.9421*0.191=761.6 e) Since there are just four years of data and most values are near 1 we cannot discern a well-defined cycle. f) Y21=220.53+19.9421*0.191=761.6 It agrees with the values computed in part c g) Excel Spreadsheet h) Prediction intervals for the next 4 quarters t=17,18,19,20 t=17:654.094,679.542 t=18:869.038,894.542 t=19:469.107,494.556 t=20:286.977,312.426 8.1) Smoothing equation l0=t=1nYtn Which is the average of the first series values lT=αyT+(1-α)lT-1 α:smoothing constant ...

Words: 415 - Pages: 2

Free Essay

Time Series Analysis

...ON TIME SERIES ANALYSIS PLANNING AND CONTROL OF OPERATONS FAWAZ MOHAMED KUTTY 215112035 MBA Ist YEAR TIME SERIES ANALYSIS A time series may be defined as a set of values of a variable collected and recorded in a chronological order of the time intervals. Time series is used by statisticians to describe the flow of economic activity. In short time series refers to the data depending on time. It refers to a set of observations concerning any activity against different periods of time. The duration of the time period may be hourly, daily, weekly, monthly or yearly. According to Morris Hamburg “A time series is a set of statistical observations arranged in chronological order”. Therefore time series is also called historical data or historical series. The study of movement of quantitative data through time is referred to as ‘time series analysis’. Time series is of great importance to the planners of economic development and economists. The success of planning depends upon making accurate forecasts of future conditions of economy. It enables the economists to foresee what is likely to come and to analyze the repercussions of past behavior. The analysis of time series enables us to understand the past behavior or performance. Time series analysis can be used to know how the data changes over time and find out probable reasons for such change. UTILITY OF TIME SERIES ANALYSIS Analysis of time series is of relevance whenever a variable is found to vary over time. Variable...

Words: 1324 - Pages: 6

Free Essay

Time Series Analysis Summary

...Time Series Analysis Summary Tokelo Khalema 2008060978 BSc. Actuarial Science University of the Free State Bloemfontein November 1, 2012 Time Series Analysis A time-series is a stochastic process {Xt : t = 1, . . . , T } with a continous state space and discrete time domain. It arises naturally as an ordered series of values observed over time. Examples include daily closing prices of a stock index recorded over several years, say, the flow rate of the River Nile, road casualties in Great Britain over the years 1969-84, etc. Stationary time-series are particularly easy to analyse. A series is stationary if its mean and variance are constant over time. Special aids are available to help determine whether or not a series is stationary. Particularly notable in this regard are the autocorrelation function (ACF) and the partial autocorrelation function (PACF). These are plots of the sample autocorrelation and partial autocorrelation coefficients at various time lags, respectively. If the ACF decays gradually to zero, then the series is non-stationary. If on the other hand the ACF and PACF decay rapidly to zero, then the series is stationary. A series being non-stationary can be brought about by, among others, a trend, irregular fluctuations, or seasonal variation. Non-constant variance, or as commonly called, heteroscedasticity can be eliminated by using a variance-stabilising transformation. A number of ways exist that eliminate a trend. Two of which are, to subtract a regression line...

Words: 1929 - Pages: 8

Free Essay

Time Series Analysis

...REPORT ON TIME SERIES ANALYSIS REPORT ON TIME SERIES ANALYSIS SUBMITTED TO M. KHAIRUL HOSSAIN PROFESSOR Department Of Finance University Of Dhaka SUBMITTED BY Group – 17 Section-A BBA 12th Batch Department Of Finance WE ARE... |Sl. No |Name |Roll No | |1. |Dulal Paul |12-143 | |2. |Rahat Hussain Md. Zaidy |12-149 | |3. |MD. Arif Hasan |12-150 | |4. |MD. Khurshid Alam |12-170 | |5. |MD. Saiful Islam |12-254 | Letter of Transmittal Date: 16th September, 2008 M. Khairul Hossain Professor Department Of Finance Faculty of Business Studies University of Dhaka Subject: Submission of report We are pleased to submit the following report on the “Time Series Analysis”. By completing the report, we have got acquainted importance and relevance of time series on business application. We also perceived idea on the whole process of Time Series Analysis. We acquired knowledge about the method...

Words: 2921 - Pages: 12

Premium Essay

Associative and Time Series Forecasting

...Forecasting Models: Associative and Time Series Forecasting involves using past data to generate a number, set of numbers, or scenario that corresponds to a future occurrence. It is absolutely essential to short-range and long-range planning. Time Series and Associative models are both quantitative forecast techniques are more objective than qualitative techniques such as the Delphi Technique and market research. Time Series Models Based on the assumption that history will repeat itself, there will be identifiable patterns of behaviour that can be used to predict future behaviour. This model is useful when you have a short time requirement (eg days) to analyse products in their growth stages to predict short-term outcomes. To use this model you look at several historical periods and choose a method that minimises a chosen measure of error. Then use that method to predict the future. To do this you use detailed data by SKU's (Stock Keeping Units) which are readily available. In TSM there may be identifiable underlying behaviours to identify as well as the causes of that behaviour. The data may show causal patterns that appear to repeat themselves – the trick is to determine which are true patterns that can be used for analysis and which are merely random variations. The patterns you look for include: Trends – long term movements in either direction Cycles - wavelike variations lasting more than a year usually tied to economic or political conditions...

Words: 1499 - Pages: 6

Free Essay

Problem -Given Is a Historical Time Series for Job

...Problem -Given is a historical time series for job Click Link Below To Buy: http://hwaid.com/shop/problem-given-historical-time-series-job/ 1. Given is a historical time series for job services demand in the prior 6 months. Month Demand 1 799 2 816 3 789 4 814 5 815 6 805 2. Use the table below to answer all questions: Month Demand Forecast 1 799 F1 2 816 F2 3 789 F3 4 814 F4 5 815 F5 6 805 F6 7 F7 3. a) The F3 by using Naïve forecasting method = _________ b) The F7 by using Naïve forecasting method = _________ 2 1. Given is a historical time series for job services demand. Period Demand Forecast 1 313 F1 2 289 F2 3 208 F3 4 325 F4 5 219 F5 6 323 F6 7 302 F7 8 299 F8 F9 F10 F11 F12 2. If you were not able to generate a forecast for particular period, you should enter N. Also, F3 means Period 3 Forecast. Use Weighted moving average with weights of 0.09, 0.11, 0.16, 0.25, 0.39 to answer Questions 1) to 3). 1) The F5 = ______ 2) The F12 = ______ (in two decimal places) 3) If you were told to prepare a forecast schedule, the forecast schedule should contain forecast demands from Period ______ to Period ______. Use Exponential Smoothing with alpha = 0.29 to answer Questions 4) to 6) 4) The F2 = ______ (in two decimal places) 5) The F3 = ______ (in two decimal places) 6) If you were told to prepare a forecast schedule, the forecast schedule should contain forecast demands from Period ______...

Words: 571 - Pages: 3

Premium Essay

A Time Series Forecasting Analysis on the Monthly Stocks of Rice in the Philippines

...A Time Series Forecasting Analysis on the Monthly Stocks of Rice in the Philippines A Research Paper Presented To Dr. Cesar Rufino Of the Department of Economics School of Economics De La Salle University, Manila In Partial Fulfillment of the Course Requirements in Economic Forecasting (ECOFORE) Term 3 AY 2014-2015 Submitted by: Jayme, Kevin Matthew D. April 24 2015 0 I. Introduction The Philippines has been the accredited as an agricultural nation that provides different types of agricultural related goods, both for the domestic and international market. Rice has been the staple food in the Philippine to 80% of the population as it is customary diet that has been in beaded in the Philippine culture (Drilon Jr., 2012). Despite the strong history of agriculture and the skills and weather condition perfect for growth of rice, decrease of land and increase of total population around the Philippines decrease the opportunity for the population to have access to rice. In addition, neighboring countries, such as Thailand and Vietnam, had been on the rise of rice exportation. Not to mention the implementation of the ASEAN integration is happening in 2015. This means that the Philippines is lagging behind as it is the 8th largest exporters of rice in the world (Tiongco & Francisco, 2011). Institution, such as International Rice Research Institute (IRRI), has gone into research and development of rice growth in different conditions and situation...

Words: 2806 - Pages: 12

Free Essay

Development of Time Series Model to Study Historical Trend of Road Traffic Accidents in the United States and Inspect the Factors Affecting the Trend

...DEVELOPMENT OF TIME SERIES MODEL TO STUDY HISTORICAL TRE ND OF ROAD TRAFFIC ACCIDENTS IN THE UNI TED STATES AND INSPECT THE FACTORS AFFECTING THE TREND Ashutosh Kedia M.Tech Project Thesis 2015 DEVELOPMENT OF TIME SERIES MODEL TO STUDY THE HISTORICAL TREND OF ROAD TRAFFIC ACCIDENTS IN THE UNITED STATES AND INSPECT THE FACT ORS AFFECTING THE TREND Thesis submitted to the Indian Institute of Technology, Kharagpur For award of the degree of Master of Technology by Ashutosh Kedia Under the guidance of Prof. Sudeshna Mitra DEPARTMENT OF CIVIL ENGINEERING INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR MAY 2015 ©2015 Ashutosh Kedia. All rights reserved. Page i M.Tech Project Thesis 2015 APPROVAL OF THE VIVA-VOCE BOARD 05/05/15 Certified that the thesis entitled DEVELOPMENT OF TIME SERIES MODEL TO STUDY HISTORICAL TREND OF ROAD TRAFFIC ACCIDENTS IN THE UNITED STATES AND INSPECT THE FACTORS AFFECTING THE TREND submitted by ASHUTOSH KEDIA to the Indian Institute of Technology, for the award of the degree, Master of Technology, has been accepted by the external examiners and that the student has successfully defended the thesis in the viva-voce examination held today. (External Examiner) (Chairman) Page ii (Supervisor) M.Tech Project Thesis 2015 CERTIFICATE This is to certify that the thesis entitled “Development of Time Series Model to Study Historical Trend of Road Traffic Accidents in the United States and Inspect the Factors Affecting the Trend” submitted by Ashutosh Kedia...

Words: 17338 - Pages: 70

Premium Essay

Forecasting with Time Series

...Forecasting with Time Series QRB/501 Quantitative Reasoning for Business February 7, 2012 Forecasting with Time Series For most companies, forecasting is very important. Their future can be determined with forecasting and this also helps pin point the problems of the past. Forecasting can be done in many methods, depending on what exactly is being forecasted. A forecasting tool used to determine demand for various commodities or goods in a given marketplace over the course of a typical year (or a shorter time period). Such an index is based on data from previous years that highlights seasonal differences in consumption. In some industries, the seasonality index experiences huge swings. (Business Dictionary, 2012) This forecasting tool is known as seasonal indexing. Find the seasonal index for summer historical inventory data below. Summer Historical Inventory Data Summer historical inventory data measures monthly figures in units four times a year. Data forecasted helps the company with inventory demands for the following year. Organizations should average a certain amount each month for a four year forecast. Organizations estimate demand for that month for the fifth year. Companies should average demands each year, and display average demands per month for that year. Data helps provide the organization, an overview with demands from one year to another year (brainmass, 2004, 2011). Organizations should identify the sequence of observation...

Words: 1074 - Pages: 5

Premium Essay

Time Series Methodologies

...Applying Time Series Methodologies Derek Griffin RES/342 March 22, 2012 Olivia Scott Applying Time Series Methodologies MEMO To: Myra Reid, VP of Production From: Derek Griffin, Market Analyst Date: 22 March 2012 Subject: Three Week Analysis Simulation to Predict Blues Inc. Forecast Message: Over the past three weeks an indebt research analysis was conducted to provide Blues Inc reasonably accurate forecast that will ensure continued growth to the six percent market share of a 45 billion dollar industry. In week one the marketing team was given a directive from the Chief Executive Officer, Barbara Baderman, to have an effective advertising strategy in place to become the industry leader. A regression analysis was performed using sales as the selected variable for the strong positive relationship to advertising budget. The correlation coefficient of sales with the advertising budget is 0.96, which was higher than the relationship of competitors advertising budget or retail coverage. Sales with a lower standard error indicate a better predicted forecast. Using the regression equation and expected sales of 2,400 million, the forecasted advertising budget should be set at 162 million. During week two the marketing team was challenged to predict the market sales for the next year. Denim sales have increased five percent over the past four years and is expected to increase again next year. The team used the weighted moving average with a weight of .9 for the...

Words: 455 - Pages: 2

Premium Essay

Time Series Analysis

...Part I Task 1 Type of Property: Bungalow Location: Taman Tun Dr. Ismail, Kuala Lumpur |Number |Square Feet |Price (RM'000) | |1 |4500 |3280 | |2 |4800 |4180 | |3 |4500 |3300 | |4 |4500 |3300 | |5 |5000 |4100 | |6 |5000 |4700 | |7 |4000 |3300 | |8 |5000 |5000 | |9 |4352 |4000 | |10 |4000 |3300 | |11 |4000 |4000 | |12 |7000 |7800 | |13 |4352 |4000 | |14 |4300 |3280 | |15 |4000 |4300 | |16 |3800 |4500 | |17 |7000 |7800 | |18 |5000 |4700 | |19 |5650 |2600 | |20 |5000 |3880 | |21 |6000 |4180 | |22 |5200...

Words: 1820 - Pages: 8

Premium Essay

Forecasting

...FORECASTING - a method for translating past experience into estimates of the future. Forecasting is the process of making statements about events whose actual outcomes (typically) have not yet been observed. A commonplace example might be estimation of the expected value for some variable of interest at some specified future date. Prediction is a similar, but more general term. Both might refer to formal statistical methods employing time series, cross-sectional or longitudinal data, or alternatively to less formal judgemental methods. Usage can differ between areas of application: for example in hydrology, the terms "forecast" and "forecasting" are sometimes reserved for estimates of values at certain specific future times, while the term "prediction" is used for more general estimates, such as the number of times floods will occur over a long period. Risk and uncertainty are central to forecasting and prediction; it is generally considered good practice to indicate the degree of uncertainty attaching to forecasts. The process of climate change and increasing energy prices has led to the usage of Egain Forecasting of buildings. The method uses Forecasting to reduce the energy needed to heat the building, thus reducing the emission of greenhouse gases. Forecasting is used in the practice of Customer Demand Planning in every day business forecasting for manufacturing companies. The discipline of demand planning, also sometimes referred to as supply chain forecasting, embraces...

Words: 3665 - Pages: 15

Free Essay

Vietnam Stock

...first success on the way of setting-up SEC in Vietnam. VN-Index of the HSEC, however, has experienced a truly ups and downs movement and changed considerable during almost last two years. In the first section on July 28, 2000, VN-Index was 100 points and increased to a peak of 571.04 points in June 25, 2001 before sliding to lower 150 points in the first months of the year 2003. Therefore, the problem is: Why they fluctuated so much, lack of orientation, and whether or not they reflected to some extent the real health of the related stock companies? From a common sense, some experts from the brokerage firms said that the stock prices at the HSEC has been fluctuated by “sheep flock effect” psychology of “naive speculators” (short time but...

Words: 3203 - Pages: 13