Financial institutions and corporations as well as individual investors and researchers often use financial time series data (such as asset prices, exchange rates, GDP, inflation and other macroeconomic indicators) in economic forecasts, stock market analysis or studies of the data itself.
But refining data is key to being able to apply it to your stock analysis. In this article, we'll show you how to isolate the data points that are relevant to your stock reports.
Cooking Raw Data
Data points are often nonstationary or have means, variances and covariances that change over time. Nonstationary behaviors can be trends, cycles, random walks or combinations of the three.
Nonstationary data, as a rule, are unpredictable and cannot be modeled or forecasted. The results obtained by using nonstationary time series may be spurious in that they may indicate a relationship between two variables where one does not exist. In order to receive consistent, reliable results, the nonstationary data needs to be transformed into stationary data. In contrast to the nonstationary process that has a variable variance and a mean that does not remain near, or returns to a longrun mean over time, the stationary process reverts around a constant longterm mean and has a constant variance independent of time.
Copryright © 2007 Investopedia.com 
Figure 1 
Types of NonStationary Processes
Before we get to the point of transformation for the nonstationary financial time series data, we should distinguish between the different types of the nonstationary processes. This will provide us with a better understanding of the processes and allow us to apply the correct transformation. Examples of nonstationary processes are random walk with or without a drift (a slow steady change) and deterministic trends (trends that are constant, positive or negative, independent of time for the whole life of the series).
Copryright © 2007 Investopedia.com 
Figure 2 
 Pure Random Walk (Y_{t} = Y_{t1} + ε_{t} )
Random walk predicts that the value at time "t" will be equal to the last period value plus a stochastic (nonsystematic) component that is a white noise, which means ε_{t }is independent and identically distributed with mean "0" and variance "σ²". Random walk can also be named a process integrated of some order, a process with a unit root or a process with a stochastic trend. It is a non mean reverting process that can move away from the mean either in a positive or negative direction. Another characteristic of a random walk is that the variance evolves over time and goes to infinity as time goes to infinity; therefore, a random walk cannot be predicted.  Random Walk with Drift (Y_{t} = α + Y_{t1 }+ ε_{t} )
If the random walk model predicts that the value at time "t" will equal the last period's value plus a constant, or drift (α), and a white noise term (ε_{t}), then the process is random walk with a drift. It also does not revert to a longrun mean and has variance dependent on time.  Deterministic Trend (Y_{t} = α + βt + ε_{t })
Often a random walk with a drift is confused for a deterministic trend. Both include a drift and a white noise component, but the value at time "t" in the case of a random walk is regressed on the last period's value (Y_{t1}), while in the case of a deterministic trend it is regressed on a time trend (βt). A nonstationary process with a deterministic trend has a mean that grows around a fixed trend, which is constant and independent of time.  Random Walk with Drift and Deterministic Trend (Y_{t} = α + Y_{t1 }+ βt + ε_{t })
Another example is a nonstationary process that combines a random walk with a drift component (α) and a deterministic trend (βt).It specifies the value at time "t" by the last period's value, a drift, a trend and a stochastic component. (To learn more about random walks and trends, see our Financial Concepts tutorial.)
Trend and Difference Stationary
A random walk with or without a drift can be transformed to a stationary process by differencing (subtracting Y_{t1 }from Y_{t, }taking the difference Y_{t}  Y_{t1}) correspondingly to Y_{t}  Y_{t1} = ε_{t }or Y_{t}  Y_{t1 }= α + ε_{t} and then the process becomes differencestationary. The disadvantage of differencing is that the process loses one observation each time the difference is taken.
Copryright © 2007 Investopedia.com 
Figure 3 
A nonstationary process with a deterministic trend becomes stationary after removing the trend, or detrending. For example, Yt = α + βt + εt is transformed into a stationary process by subtracting the trend βt: Yt  βt = α + εt, as shown in Figure 4 below. No observation is lost when detrending is used to transform a nonstationary process to a stationary one.
Copryright © 2007 Investopedia.com 
Figure 4 
In the case of a random walk with a drift and deterministic trend, detrending can remove the deterministic trend and the drift, but the variance will continue to go to infinity. As a result, differencing must also be applied to remove the stochastic trend.
Conclusion
Using nonstationary time series data in financial models produces unreliable and spurious results and leads to poor understanding and forecasting. The solution to the problem is to transform the time series data so that it becomes stationary. If the nonstationary process is a random walk with or without a drift, it is transformed to stationary process by differencing. On the other hand, if the time series data analyzed exhibits a deterministic trend, the spurious results can be avoided by detrending. Sometimes the nonstationary series may combine a stochastic and deterministic trend at the same time and to avoid obtaining misleading results both differencing and detrending should be applied, as differencing will remove the trend in the variance and detrending will remove the deterministic trend.

Mutual Funds & ETFs
ETF Analysis: First Trust Dow Jones Global Sel Div
Find out about the First Trust Dow Jones Global Select Dividend Index Fund, and learn detailed information about characteristics and suitability of the fund. 
Fundamental Analysis
What Causes Inflation in the United States
Inflation is the main catalyst behind U.S monetary policy. But what causes this phenomenon of sustained rising prices? Read on to find out. 
Term
What are Mutually Exclusive Events?
In statistics, mutually exclusive situations involve the occurrence of one event that does not influence or cause another event. 
Mutual Funds & ETFs
ETF Analysis: PowerShares DB Commodity Tracking
Find out about the PowerShares DB Commodity Tracking ETF, and explore a detailed analysis of the fund that tracks 14 distinct commodities using futures contracts. 
Mutual Funds & ETFs
ETF Analysis: PowerShares FTSE RAFI US 1000
Find out about the PowerShares FTSE RAFI U.S. 1000 ETF, and explore detailed analysis of the fund that invests in undervalued stocks. 
Mutual Funds & ETFs
ETF Analysis: Vanguard IntermediateTerm Corp Bd
Learn about the Vanguard IntermediateTerm Corporate Bond ETF, and explore detailed analysis of the fund's characteristics, risks and historical statistics. 
Mutual Funds & ETFs
Top 3 Switzerland ETFs
Explore detailed analysis and information of the top three Swiss exchangetraded funds that offer exposure to the Swiss equities market. 
Economics
The Problem With Today’s Headline Economic Data
Headwinds have kept the U.S. growth more moderate than in the past–including leverage levels and an aging population—and the latest GDP revisions prove it. 
Economics
Explaining the Participation Rate
The participation rate is the percentage of civilians who are either employed or unemployed and looking for a job. 
Fundamental Analysis
Is India the Next Emerging Markets Superstar?
With a shift towards manufacturing and services, India could be the next emerging market superstar. Here, we provide a detailed breakdown of its GDP.

Cost Accounting
A type of accounting process that aims to capture a company's ... 
ZeroSum Game
A situation in which one person’s gain is equivalent to another’s ... 
Supply
A fundamental economic concept that describes the total amount ... 
PrincipalAgent Problem
The principalagent problem develops when a principal creates ... 
Discount Bond
A bond that is issued for less than its par (or face) value, ... 
Compound Annual Growth Rate  CAGR
The Compound Annual Growth Rate (CAGR) is the mean annual growth ...

What assumptions are made when conducting a ttest?
The common assumptions made when doing a ttest include those regarding the scale of measurement, random sampling, normality ... Read Full Answer >> 
What is the utility function and how is it calculated?
In economics, utility function is an important concept that measures preferences over a set of goods and services. Utility ... Read Full Answer >> 
What are some of the more common types of regressions investors can use?
The most common types of regression an investor can use are linear regressions and multiple linear regressions. Regressions ... Read Full Answer >> 
What types of assets lower portfolio variance?
Assets that have a negative correlation with each other reduce portfolio variance. Variance is one measure of the volatility ... Read Full Answer >> 
When is it better to use systematic over simple random sampling?
Under simple random sampling, a sample of items is chosen randomly from a population, and each item has an equal probability ... Read Full Answer >> 
What are some common financial sampling methods?
There are two areas in finance where sampling is very important: hypothesis testing and auditing. The type of sampling methods ... Read Full Answer >>