Financial Time Series Analysis for Trading
₹6500/-₹25999/-
75% OFF
Get for ₹5525 with Course Bundle
- Live Trading
- Learning Track
- Prerequisites
- Syllabus
- About author
- Testimonials
- Faqs
Learn Financial Time Series
- Explain the basics of time series and its components
- Learn about simple, cumulative and log returns.
- Apply linear and multivariate regression on stock prices.
- Apply correlation analysis on various securities and explain the intuition behind ACF and PACF, as well as plot them in python.
- Explain types of noise and residuals. Define stationarity and list its properties.
- Describe and implement AR, MA, ARMA, ARIMA, SARIMA, ARCH and GARCH models.
- Explain the importance of volatility and its stylised facts. Learn about the limitations of time series analysis and learn how to enhance your model.
- Backtest and live trade time series models on financial markets data and implement all the concepts learned in a capstone project.

Skills for Financial Time Series
Statistics
- Log Returns
- Linear Regression
- Correlation
- PACF, ACF, AIC and BIC
- Goodness of fit, R-Squared
Time Series Models
- AR, MA
- ARMA
- ARIMA
- SARIMA
- ARCH, GARCH
Python
- Numpy
- Pandas
- Matplotlib
- Statsmodel

learning track 1
This course is a part of the Learning Track: Algorithmic Trading for Beginners
Course Fees
Full Learning Track
These courses are specially curated to help you with end-to-end learning of the subject.
Course Features
- Community
Faculty Support on Community
- Interactive Coding Exercises
Interactive Coding Practice
- Capstone Project
Capstone Project Using Real Market Data
- Trade & Learn Together
Trade and Learn Together
- Get Certified
Get Certified
Prerequisites
This course is recommended if you are looking to learn and implement time series analysis in live trading. This course requires basic familiarity with Python packages such as NumPy and Pandas to easily follow the Python codes covered in the course. If you want to learn NumPy and Pandas, then you can enroll in the 'Python For Trading' course on Quantra.
Financial Time Series Course
- IntroductionUnderstand how one can implement math and statistics in financial markets to create trading strategies. This section also gives an overview of different types of learning units which comprise the course: videos, coding exercises, Python notebooks, quizzes and trading platform integrated units. You would also be taken through the course syllabus which is perfect for beginners in the domain. This entire section is available for free preview.
What is Time Series?
A time series should be recorded periodically and without any gaps. In this section learn what constitutes a time series and what does not. For example, new iPhone model announcements are not periodical and hence are not fit for time series analysis. Further, understand when a time-series analysis should be carried out.Introduction to Time Series2m 9sFinding the Time Series2mFrequency in Time Series Analysis2mData Set in Time Series2mAcquisitions in Time Series2mWhy is Time Series Analysis Required?2m 51sReasons for Time Series Analysis2mWhen Is Time Series Analysis Not Required?1m 42sScenario for Not Analysing Time Series2mSimple and Cumulative Returns
Perfect section for beginners in finance to get started with. Returns are simply the change in the prices from yesterday. Learn how returns are calculated and further, why simple returns should be multiplied and not added, to give cumulative returns.Introduction to Returns2m 40sAnalysing Returns or Price2mSimple Daily Returns2mTotal Returns2mCumulative Returns2m 22sSimple or Cumulative Returns2mHow to Use Jupyter Notebook?1m 54sCalculating Returns and Cumulative Returns10mFrequently Asked Questions10mFord Cumulative Returns2mLog Returns
Often a confusing concept, we have made log returns intuitive and easier for you to understand. Sometimes a price series changes drastically over a course of time. To get a clear picture of how a time series changes, you use the log prices. Calculate its returns to get the log returns.Log Prices3m 3sLog Price Graphs2mAdvantage of Log Prices2mPercentage Change in Price2mTotal Log Returns2mLog Returns2m 10sSimple or Log Returns2mDaily Log Returns2mSignificance of Log Price2mSignificance of Log Returns2mLog Prices or Daily Log Returns2mLog Returns10mCalculating Total Log Returns2mPortfolio Returns Calculations10mPortfolio Log Returns2mPortfolio Returns2mAdditional Reading10m- Components of Time SeriesBefore you implement a financial time series model, you need to understand the different components present in the time series. Learn how to identify components such as trend, mean reversion, seasonality and cyclicality using real market data.Components of Time Series1m 34sIdentifying Time Series Components2mTrending Time Series2m 26sMean Reverting Time Series1m 15sTrending or Mean Reverting2mTrending and Mean Reverting Simultaneously2mCyclical Time Series2mReason for Cyclicality2mTrending or Cyclical2mIdentifying Cyclical Industries2mSeasonal Time Series2m 27sImportance of Seasonality2mIdentifying Seasonal Pattern2mCyclical or Seasonal Time Series2mSeasonal and Trending Simultaneously2m
- Linear RegressionTake your first step in predicting stock prices! If two assets move generally in the same direction at the same time, you can use one asset’s time series to forecast the other. This method is called linear regression. Learn the basic principles of linear regression in this section.Linear Regression Fundamentals3m 16sNecessity of Linear Regression2mPredicting Output With Zero Slope2mGraph of Linear Regression2mLinear Relationship2mScatter Plot2mLinear Regression Model10mPrint Model Summary5m
- Types of ErrorsPredictions are erroneous! This section details the different types of error calculations. Depending on the linear regression model, there will be data points which lie outside the line. The distance between the fitted line and the datapoint is called error. Learn to understand errors and improve your prediction models going ahead.Types of Error Calculations3m 6sIssues With Sum of Errors2mMean Squared Error Calculation2mInferring Sum of Squared Errors2mAdvantage of Mean Squared Error2mBenefit of SSE Over SAE2mTypes of Errors10mCalculate the Error5mCalculate Mean Absolute Error5mCalculate Mean Squared Error5mError and Outliers2mCalculate Root Mean Squared Error5mRoot Mean Squared Error of a Straight Line2mWhich is a better RMSE?2mCalculate Mean Absolute Percentage Error5m
- Goodness of FitGoodness of fit is used as a criteria for measuring the fitted line’s effectiveness. This section tells us how the goodness of fit is calculated. This will help you judge whether your prediction of the asset is good enough to be converted into a trading strategy.Introduction to Goodness of Fit4m 18sWhy Goodness of Fit2mError of a Good Model2mR-Squared Value2mResidual Plot2mPattern in Residual Plot2mHigh R-Squared Value2mR-Squared10mCalculate R-Squared5mLimitations of R-Squared2mAssumptions for Linear Regression10mHighest R-Squared2mLinearity2mNot an Assumption for Linear Regression2mAutocorrelation2mResiduals2m
Multivariate Linear Regression
In this section, you will learn how multiple independent variables can be used to forecast the position of the dependent variable with a live example. You will use two stocks such as Bank of America and Citigroup to forecast the price of a third stock, J.P. Morgan. You will also look at the limitations of the linear regression approach. Depending on the relation between the assets, you can use multiple assets to predict one asset’s price.Multivariate Linear Regression1m 46sSelecting Multiple Variables in Model2mMultivariate Linear Regression Equation2mRequirement of Multivariate Linear Regression2mLagged Version of Own Time Series2mMultivariate Linear Regression Model10mOutput of Multivariate Regression Model2mLimitations and Advantages of Linear Regression2m 39sEliminating Outliers2mLinear and Non-linear Models2mLimitation of Linear Models2mAdditional Reading10m- Correlation AnalysisA widely used statistical concept in finance, correlation analysis helps in establishing a possible relationship between security prices. This in-turn helps in predicting the future price of securities. In this section, you will learn about correlation,how it is different from covariance and how to calculate covariance and correlation coefficients with their limitations.Correlation and Covariance4m 38sCalculation of Covariance and Correlation10mCovariance Value2mCorrelated Assets2mDirection of the Linear Relationship2mPortfolio Diversification2mImplementation of Correlation Coefficient10mAre Numerical Calculations Exact?10mCalculate the Correlation Coefficient5mCalculate the Rolling Correlation5m
Autocorrelation and Partial Autocorrelation
These two concepts go hand in hand while modelling time series, autocorrelation and partial autocorrelation. In this section learn about the intuition of ACF and PACF, their differences and how to plot ACF & PACF along their application.What is Autocorrelation?4m 26sAutocorrelation2mAutocorrelation Value2mStatistically Significant Values2mInterpretation of ACF plot2mForecasting Using ACF Plot2mWhat is Partial Autocorrelation?4mPartial Autocorrelation2mPACF Value2mModel Formation Using PACF Plot2mACF and PACF Plotting in Python10mSignificance of Blue Region in ACF and PACF2mACF Plot in Python5mPACF Plot in Python5mAdditional Reading10m- NoiseThe fourth component of time series, noise, can be caused by either a glitch in the recording process or a temporary deviation from the system. White noise in particular, is the noise remaining after the time series model has been optimised. The presence of white noise in the model indicates that our time series model cannot be optimised further.Noise3m 34sSimilarity of Returns and White Noise2mStationarity in White Noise2mError Plot in White Noise2mAdditional Reading10m
Autoregressive Model
Autoregressive model is based on the linear regression model which assumes past values of a time series have ability to predict future values. In this section, you will learn all about intuition of the autoregressive model, its equation and how to find the optimal lag term for an autoregressive model.Overview of Part II2m 50sAutoregressive Model - I2m 40sWhat is an Autoregressive Model?2mRepresentation of AR Model2mAutoregressive Model - II3m 9sPrerequisite for AR Model2mOrder of AR Model2mWhich AR Model to Use?2mLimitations of Using More Lag Terms2mImplement Autoregressive Model
By now, you will already be familiar with the AR model. In this section, you will learn to implement it on the financial time series along with the Python packages used for time series forecasting.Simple AR Model10mShift in Predicted Price5mAR Model of Order p10mTrain an AR Model of Order p2mAdditional Reading10m- Moving Average ModelThe second model in the family of time series models assumes that past error terms have the ability to predict future values. Learn about intuition of MA model, its equation and how to find the optimal lag term for an autoregressive model.Moving Average Model4m 47sWhat is Moving Average Model?2mPredict Price Using MA Model2mWhich MA Model to Use?2mWhat is Definition of MA(2) Model?2mSimple MA Model10mMA Model of Order q10mGenerate a Trading Signal2mAdditional Reading10m
- ARMAARMA is a more advanced model than AR and MA models. In this section, you will learn about the ARMA model, and its equation.ARMA Model2m 11sWhat is ARMA Model?2mEquation of ARMA Model2mWhich ARMA Model to Use?2mCaveats of AR, MA and ARMA10m
- StationarityA stationary time series is supposed to have constant mean and variance, irrespective of time period. The AR, MA and ARMA models require stationarity for their smooth functioning. In this section, you will learn how to convert a non-stationary financial time series to a stationary time series.Stationarity5m 19sStationarity in Kodak Prices2mStationarity and Zero Mean2mChanging Variance and Constant Mean2mSingle Order Differencing2mDefinition of Stationarity2mIdentifying Graphs of Stationarity2mIdentify the Stationary Series2mADF Test10mInterpret the Output of ADF Test2mApply the ADF Test on a Given Series5mConvert Non Stationary Series to a Stationary5mApply the ADF Test on the Difference of the Series5mAdditional Reading10m
- ARIMA ModelA forecasting model that works with non stationary time series. In this section, you will learn about the ARIMA model and its equation. This section also covers finding the optimal order for each parameter associated with the ARIMA model. Finally, you will learn to make a prediction using the ARIMA model.ARIMA Model3m 24sNeed of ARIMA Model2mOrder of Integrated2mPredict Price from Change in Price2mARMA Model on Non Stationary Data2mWhich ARIMA Model to Use?2mEquation of an ARIMA Model10mAIC and BIC10mGetting Started with ARIMA Model10mShould you Apply ARIMA?2mARIMA Model of Order (p, d, q)10mCorrect Order of ARIMA2mOptimal p for ARIMA2mAdditional Reading10mBest ARIMA Model Selection10m
- Live Trading on BlueshiftThis section will walk you through the steps involved in taking your trading strategy live. You will learn about backtesting and live trading platform, Blueshift. You will learn about code structure, various functions used to create a strategy and finally, paper or live trade on Blueshift.Uninterrupted Learning Journey with Quantra2mSection Overview2m 19sLive Trading Overview2mVectorised vs Event Driven2mProcess in Live Trading2mReal-Time Data Source2mBlueshift Code Structure2m 57sImportant API Methods10mSchedule Strategy Logic2mFetch Historical Data2mPlace Orders2mBacktest and Live Trade on Blueshift4m 5sAdditional Reading10mBlueshift Data FAQs10m
- Live Trading TemplatePaper/Live Trading ARIMA Strategy10mFAQs for Live Trading on Blueshift10m
- SARIMA ModelSARIMA or Seasonal ARIMA is one step ahead forecasting model than ARIMA. In this section, you learn about SARIMA, and why it is a must-know model in a time series forecasting. This section also covers the representation of SARIMA and how to find the optimal order of its parameters.SARIMA Model10mNeed of SARIMA Model2mOrder of m in SARIMA2mOrder of D in SARIMA2m
Introduction to Volatility
A time series is highly volatile if the price series keeps changing drastically within the time period. In this section, you will learn the definition of volatility and calculate daily and annualised volatility.Fundamentals of Volatility2m 33sDefinition of Volatility2mSteps in Volatility Calculations2mImportance of Volatility3m 12sImpact of Events on Volatility2mImpact of Recession on Volatility2mHalf-yearly Volatility2mCalculate Volatility10mData to Calculate Volatility2mWhy Log Returns While Calculating Volatility?2mResample Daily Data to Monthly Period5mCalculate the Monthly Volatility5mConvert the Monthly Volatility to Annualised Volatility5mHalf-Yearly Volatility to Annualised Volatility2m- Stylised Facts and Importance of VolatilityIn this section, you will learn the stylised facts of volatility such as volatility clustering, mean-reversion, long memory and leverage effect. You will also learn about its importance as a risk indicator and how it can be used to trade the volatility index (VIX)Stylised Facts of Volatility3m 44sSelecting Stylised Facts2mMean-reverting Behaviour of Volatility2mReturns and Volatility2mReason for Asymmetry in Volatility2mApplications of Volatility3m 23sDifferentiating Historical and Implied Volatility2mCalculating Implied Volatility2mImplied Volatility and SP5002mTrading VIX2mUsage of Volatility2m
ARCH
The volatility of the real world financial data usually changes with time. Learn how to use the ARCH model to predict the volatility of an asset using its past returns.Need for the ARCH and GARCH model1m 45sEffect of Returns on Volatility2mModel Selection: Volatility in a Range2mModel Selection: Volatility Not in a Range2mIntroduction to the ARCH Model3m 5sComponents of the ARCH Model2mReturns-Volatility Relationship2mEquation of the ARCH Model1m 25sDerivation of the ARCH Model10mIdentify the Correct ARCH Equation2mImplementation of the ARCH Model10mDefine the ARCH Model2mForecasting using the ARCH Model2mPerformance Analysis of the ARCH Model2m 42sSummarise the ARCH Model2mCompute the ARCH Volatility2mAdditional Reading10mGARCH
The ARCH model is generalised further to include the asset returns and past volatility. Learn how to use the GARCH model to predict the volatility and then use the prediction in a trading strategy.Implementation of the GARCH Model3m 31sVolatility Clustering2mComponents of the GARCH Model2mImplementation of the GARCH Model10mFinding the Optimal Lag2mPerformance Analysis of the GARCH Model1m 50sSummarise the GARCH Model2mCompute the GARCH Volatility2mTrading Strategy using the GARCH Model10mGenerate the Strategy Signal2mAdditional Reading10m- Capstone ProjectIn this section, you will undertake a capstone project on real-world data. This project will require you to apply and practice the concepts learnt throughout this course.Capstone Project: Getting Started10mProblem Statement10mFrequently Asked Questions10mTemplate Code Files2mWorking With Pickle File5mModel Solution: TSA Capstone Project10mCapstone Solution Downloadable2m
- LimitationsTime series analysis has a wide variety of applications and acceptance in the quant community, it is not perfect. In this section, you will go through the limitations inherent in time series and also ways to overcome them.Limitations of Time Series Analysis3m 51sLimitations in Methodology2mImpact of Events on Price Data2mAdditional Reading10m
- Future EnhancementsIn this section, you will look at the present scenario of time series analysis and go through a few pointers on how to improve our existing time series models.Future Enhancements3m 31sExogenous and Endogenous Variables2mIdentifying X in SARIMAX2mStylised Fact of Volatility in EGARCH2mWorking of Pairs Trading2mIdentifying Additive Trends2mApplying Exponential Models2mAdditional Reading10m
- Automate Trading Strategy Using IBridgePyAdditional Reading10mSample Strategy to Run on Interactive Brokers2m
- Run Codes Locally on Your MachineLearn to install the Python environment in your local machine.Python Installation Overview2m 18sFlow Diagram10mInstall Anaconda on Windows10mInstall Anaconda on Mac10mKnow your Current Environment2mTroubleshooting Anaconda Installation Problems10mCreating a Python Environment10mChanging Environments2mQuantra Environment2mTroubleshooting Tips For Setting Up Environment10mHow to Run Files in Downloadable Section?10mTroubleshooting For Running Files in Downloadable Section10m
- Course SummaryThis section will give a brief summary of the course and the different concepts and models you have worked on in this course. All the codes and data files are also available as a zip file in this section.Course Summary2m 52sPython Code and Data2m
Registered Successfully!
You will receive webinar joining details on your registered email
Would you like to start learning immediately?
about author


Why quantra®?
- More in Less Time
Gain more in less time
- Expert Faculty
Get taught by practitioners
- Self-paced
Learn at your own pace
- Data & Strategy Models
Get data & strategy models to practice on your own
Faqs
- When will I have access to the course content, including videos and strategies?
You will gain access to the entire course content including videos and strategies, as soon as you complete the payment and successfully enroll in the course.
- Will I get a certificate at the completion of the course?
Yes, you will be awarded with a certification from QuantInsti after successfully completing the online learning units.
- Are there any webinars, live or classroom sessions available in the course?
No, there are no live or classroom sessions in the course. You can ask your queries on community and get responses from fellow learners and faculty members.
- Is there any support available after I purchase the course?
Yes, you can ask your queries related to the course on the community: https://quantra.quantinsti.com/community
- What are the system requirements to do this course?
Fast-speed internet connection and a browser application are required for this course. For best experience, use Chrome.
- What is the admission criteria?
There is no admission criterion. You are recommended to go through the prerequisites section and be aware of skill sets gained and required to learn most from the course.
- Is there a refund available?
We respect your time, and hence, we offer concise but effective short-term courses created under professional guidance. We try to offer the most value within the shortest time. There are a few courses on Quantra which are free of cost. Please check the price of the course before enrolling in it. Once a purchase is made, we offer complete course content. For paid courses, we follow a 'no refund' policy.
- Is the course downloadable?
Some of the course material is downloadable such as Python notebooks with strategy codes. We also guide you how to use these codes on your own system to practice further.
- Can the python strategies provided in the course be immediately used for trading?
We focus on teaching these quantitative and machine learning techniques and how learners can use them for developing their own strategies. You may or may not be able to directly use them in your own system. Please do note that we are not advising or offering any trading/investment services. The strategies are used for learning & understanding purposes and we don't take any responsibility for the performance or any profit or losses that using these techniques results in.
- I want to develop my own algorithmic trading strategy. Can I use a Quantra course notebook for the same?
Quantra environment is a zero-installation solution to get beginners to start off with coding in Python. While learning you won't have to download or install anything! However, if you wish to later implement the learning on your system, you can definitely do that. All the notebooks in the Quantra portal are available for download at the end of each course and they can be run in the local system just the same as they run in the portal. The user can modify/tweak/rework all such code files as per his need. We encourage you to implement different concepts learnt from different learning tracks into your trading strategy to make it more suited to the real-world scenario.
- If I plug in the Quantra code to my trading system, am I sure to make money?
No. We provide you guidance on how to create strategy using different techniques and indicators, but no strategy is plug and play. A lot of effort is required to backtest any strategy, after which we fine-tune the strategy parameters and see the performance on paper trading before we finally implement the live execution of trades.
- What is a capstone project? Who will review my work? What to do if I'm stuck?
What is a capstone project?
A capstone project is a multi-faceted hands-on project that lets you apply what you\'ve learned in an advanced course. This project is relevant to how the concepts are applied in real-world scenarios and helps to cement your understanding of the concepts.
Who will review my work?
Since the capstone project involves trading on real-world assets, the best measure for performance is the profitability and the performance metrics like the Sharpe ratio, drawdown, etc. There is no formal or peer review and you can refer to the model solution at the end of this section.
Will I get a solution for the above project?
We have provided a model solution to help you understand how to get things working. The input features or the hyperparameters can all be changed, added, or removed as per your need.
What to do if I\'m stuck?
If you get stuck, we encourage you to view the model solution and compare it with your solution to see what went wrong. If you still need help, feel free to reach out to us via the community forum and we will be happy to help you!
- What does "lifetime access" mean?
Lifetime access means that once you enroll in the course, you will have unlimited access to all course materials, including videos, resources, readings, and other learning materials for as long as the course remains available online. There are no time limits or expiration dates on your access, allowing you to learn at your own pace and revisit the content whenever you need it, even after you've completed the course. It's important to note that "lifetime" refers to the lifetime of the course itself—if the platform or course is discontinued for any reason, we will inform you in advance. This will allow you enough time to download or access any course materials you need for future use.
- What is a time series analysis?
Time series analysis is the process of analysing a time series data.
A time series data is the data recorded at a certain interval of time and ordered chronologically. These data are recorded from the same source and can be recorded on a certain frequency like yearly, monthly or daily. The frequency can be varied based on need or interest.
There are four components of a time series data - seasonality, trend, cyclicality and variation. Using time series analysis, you can analyse the data to understand the actual condition through the graphical and numerical representation. This will further enable you to make an informed prediction of future values.
Thus a time series analysis helps you to understand the behaviour of the past, monitor the present by tracking the changes over time and predict the future. - How does time series analysis work?
As the name suggests, time series analysis works on time-based data (data collected at a certain frequency). Time series analysis identifies the serial correlation or autocorrelation and derives the hidden insights to make future predictions.
Algorithms like autoregression and moving average method assume that the statistical properties of the series, like mean and variance, remains constant across time. This property is called Stationarity. Most of the statistical tests rely on inferences drawn from a stationary series. However, in real practice, most of the data are non-stationary in nature. In such cases, algorithms like ARIMA are used. The time series models first convert the non-stationary data into stationary data before performing further analysis. - What are the types of time series?
There are two types of time series data, regular time series and irregular time series.
Regular time series data are read at regular time intervals, like recording the temperature every hour, or average daily temperature. The data points are evenly spaced across the time span it is representing.
Irregular time series data are recorded at irregular time intervals, like withdrawal at an ATM. These data points are unevenly spaced across the time span it is representing.
These data are recorded against a timestamp and stored in chronological order. A new data point is always appended to the existing data. - How much data do you need to forecast?
The amount of data needed to forecast the future values would depend on the approach that you will use for modelling. Theoretically, the only limit is that the number of data points should be more than the number of parameters. However, in practice, you would require much more data points than that. Insufficient data might be misleading in actual practice. Forecasting results on insufficient data can ignore the underlying variability in the data and overlook the number of parameters to be estimated.
At the same time, you also need to be careful that the model might not perform very well on very old data. The difference between the actual values and values predicted by the model trained on large datasets might be very different. Thus a rolling window would be more appropriate than an expanding window if the recent lags (say, values upto lag t) have more predicting power than the older lag values (say, values from lag t+1). - What are the advantages of time series analysis?
Time series analysis is the most reliable method when it comes to analysing data recorded against time. It is useful in estimating trends and patterns in the data. Time series analysis is also very useful to measure the temporal effect.
While other methods like supervised learning techniques might fail if the data does not have enough variables, time series models forecast future values even with fewer variables/data.
The returns of an asset are volatile in nature. Time series models like ARCH and GARCH are very useful in predicting the volatility of a series. - What are the limitations of time series analysis?
Time series analysis assumes that the past behaviour of the data will continue indefinitely and using historical data to make predictions would lead to valid conclusions. It overlooks the external factors affecting the data.
Determining the appropriate model for the data is not very well defined. - What are the applications of time series analysis in financial markets?
Time series analysis is widely used to forecast the patterns in the financial markets. Time series analysis is also used to analyse how an asset or security changes over a period of time.
- What are some popular time series models?
AR, MA, ARMA and ARIMA are some of the most popular time series models. However, each technique has its own shortcomings. As you move further, the complexity of the model increases. ARIMA is one of the most used algorithms in time series analysis/forecasting, however, it fails when the data has a seasonal component.
- What is the ARIMA model used for?
ARIMA stands for Autoregressive Integrated Moving Average model. Algorithms in time series analysis like autoregressive (AR) and moving averages (MA) assume the data to be stationary. However, in real life, almost all the data are non-stationary in nature. Hence the predictions made using these algorithms might not be accurate. What an ARIMA model does is, it converts the non-stationary series into a stationary series and then applies the AR and MA model. The parameter required to convert the non-stationary series into a stationary series is the I in ARIMA.