Regression Analysis - Data Science Central2021-04-23T10:48:53Zhttps://www.datasciencecentral.com/forum/topics/regression-analysis?commentId=6448529%3AComment%3A696364&feed=yes&xn_auth=no(Multiple) Linear Regression…tag:www.datasciencecentral.com,2018-02-26:6448529:Comment:6980822018-02-26T17:16:08.395ZWayne G Fischerhttps://www.datasciencecentral.com/profile/WayneGFischerPhD
<p>(Multiple) Linear Regression does *not* require a linear relationship between the response and the predictor(s). "Linear" refers to the requirement that the model being considered is linear in the *parameters* to be estimated.<br></br> <br></br> <cite>Mithun Alva said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis#6448529Comment696275"><div><div class="xg_user_generated"><p>Hi Prashanth.</p>
<p></p>
<p>You can build a basic multiple regression…</p>
</div>
</div>
</blockquote>
<p>(Multiple) Linear Regression does *not* require a linear relationship between the response and the predictor(s). "Linear" refers to the requirement that the model being considered is linear in the *parameters* to be estimated.<br/> <br/> <cite>Mithun Alva said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis#6448529Comment696275"><div><div class="xg_user_generated"><p>Hi Prashanth.</p>
<p></p>
<p>You can build a basic multiple regression model after creating dummy variables to represent the variables with the categorical values.</p>
<p></p>
<p>Once you have built this base model, you will absolutely need to check for the following assumptions and make sure they hold. You can do this by plotting and visual inspection.</p>
<p>1) There is a linear relationship between your response and your predictors.</p>
<p>2) Your predictors are not highly correlated with each other.</p>
<p>3) The mean of the residuals (i.e. Predicted-Actual) is zero.</p>
<p>4) The residuals display constant variance</p>
<p>5) The residuals are normally distributed.</p>
<p></p>
<p>If the assumptions are not satisified, you will need to look into transforming your predictors( i.e. change x to x-squared, 1/x, log(x), (x-mean)/(standard deviation),etc.) and rechecking the assumptions. </p>
<p></p>
<p></p>
<p></p>
</div>
</div>
</blockquote> You could use a normal Linear…tag:www.datasciencecentral.com,2018-02-26:6448529:Comment:6979662018-02-26T08:37:38.852ZRamkumar Mambakkamhttps://www.datasciencecentral.com/profile/RamkumarMambakkam
<p>You could use a normal Linear Regression, as your dependent variable is continuous. The categorical variables, can be converted to dummy variables ( if there are not many unique ones ). </p>
<p>Depending on the Overfitting or underfitting, you can play on the Independent variables exclusion ( removing the greatest P-valued variables).</p>
<p>You can try out the Lazzo or Ridge models, if you want to control the coefficients.</p>
<p>You could use a normal Linear Regression, as your dependent variable is continuous. The categorical variables, can be converted to dummy variables ( if there are not many unique ones ). </p>
<p>Depending on the Overfitting or underfitting, you can play on the Independent variables exclusion ( removing the greatest P-valued variables).</p>
<p>You can try out the Lazzo or Ridge models, if you want to control the coefficients.</p> Hi Prashanth
As an alternate…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6963642018-02-22T22:38:18.190ZSagar Dawdahttps://www.datasciencecentral.com/profile/SagarDawda
<p>Hi Prashanth</p>
<p>As an alternate approach you could try using a CART (Classification And Regression Tree) like Random Forest . This can be helpful especially in circumstances when the number of categories are more in your categorical feature. So let's say if you had 25 categories then using dummy would add 25 columns. In such a scenario you'd be trapped in the curse of dimensionality. </p>
<p>Hi Prashanth</p>
<p>As an alternate approach you could try using a CART (Classification And Regression Tree) like Random Forest . This can be helpful especially in circumstances when the number of categories are more in your categorical feature. So let's say if you had 25 categories then using dummy would add 25 columns. In such a scenario you'd be trapped in the curse of dimensionality. </p> Sounds great Prashanth. Were…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6963632018-02-22T22:15:18.307ZMithun Alvahttps://www.datasciencecentral.com/profile/MithunAlva
<p>Sounds great Prashanth. Were you able to check for the above assumptions?<br></br> <br></br> <cite>Prashanth Southekal, PhD said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis?commentId=6448529%3AComment%3A696534&xg_source=msg_com_forum#6448529Comment696534"><div><div class="xg_user_generated"><p>Thanks Mithun for your help. Creating dummy variables was the key as I did not know that before. I was able to get this done and come up with a…</p>
</div>
</div>
</blockquote>
<p>Sounds great Prashanth. Were you able to check for the above assumptions?<br/> <br/> <cite>Prashanth Southekal, PhD said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis?commentId=6448529%3AComment%3A696534&xg_source=msg_com_forum#6448529Comment696534"><div><div class="xg_user_generated"><p>Thanks Mithun for your help. Creating dummy variables was the key as I did not know that before. I was able to get this done and come up with a prediction model.</p>
</div>
</div>
</blockquote> Thanks Mithun for your help.…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6965342018-02-22T22:12:23.680ZPrashanth Southekal, PhDhttps://www.datasciencecentral.com/profile/PrashanthSouthekal
<p>Thanks Mithun for your help. Creating dummy variables was the key as I did not know that before. I was able to get this done and come up with a prediction model.</p>
<p>Thanks Mithun for your help. Creating dummy variables was the key as I did not know that before. I was able to get this done and come up with a prediction model.</p> Hi Prashanth.
You can build…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6962752018-02-22T22:02:42.933ZMithun Alvahttps://www.datasciencecentral.com/profile/MithunAlva
<p>Hi Prashanth.</p>
<p></p>
<p>You can build a basic multiple regression model after creating dummy variables to represent the variables with the categorical values.</p>
<p></p>
<p>Once you have built this base model, you will absolutely need to check for the following assumptions and make sure they hold. You can do this by plotting and visual inspection.</p>
<p>1) There is a linear relationship between your response and your predictors.</p>
<p>2) Your predictors are not highly correlated with…</p>
<p>Hi Prashanth.</p>
<p></p>
<p>You can build a basic multiple regression model after creating dummy variables to represent the variables with the categorical values.</p>
<p></p>
<p>Once you have built this base model, you will absolutely need to check for the following assumptions and make sure they hold. You can do this by plotting and visual inspection.</p>
<p>1) There is a linear relationship between your response and your predictors.</p>
<p>2) Your predictors are not highly correlated with each other.</p>
<p>3) The mean of the residuals (i.e. Predicted-Actual) is zero.</p>
<p>4) The residuals display constant variance</p>
<p>5) The residuals are normally distributed.</p>
<p></p>
<p>If the assumptions are not satisified, you will need to look into transforming your predictors( i.e. change x to x-squared, 1/x, log(x), (x-mean)/(standard deviation),etc.) and rechecking the assumptions. </p>
<p></p>
<p></p>
<p></p> Read about ANCOVA model. You…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6963602018-02-22T21:55:06.685Zshaikh mohammed sabirhttps://www.datasciencecentral.com/profile/shaikhmohammedsabir
<p>Read about ANCOVA model. You can handle categorical variable using dummy variable as said by Vincent above.</p>
<p>Read about ANCOVA model. You can handle categorical variable using dummy variable as said by Vincent above.</p> It's completely depend upon y…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6963032018-02-22T04:12:48.656ZSayan Nandyhttps://www.datasciencecentral.com/profile/SayanNandy
<p>It's completely depend upon your objectives.As your dependent variable is continuous, I think you should use Multiple Linear regression Model .</p>
<p>It's completely depend upon your objectives.As your dependent variable is continuous, I think you should use Multiple Linear regression Model .</p> Thanks for your help Vincent…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6960982018-02-22T02:01:47.162ZPrashanth Southekal, PhDhttps://www.datasciencecentral.com/profile/PrashanthSouthekal
<p>Thanks for your help Vincent<br></br> <br></br> <cite>Vincent Granville said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis?commentId=6448529%3AComment%3A695879&xg_source=msg_com_forum#6448529Comment695879"><div><div class="xg_user_generated"><p>You could turn your categorical variables into dummy binary variables (google "dummy variable"), and then use any standard regression. This is often done in the context of logistic…</p>
</div>
</div>
</blockquote>
<p>Thanks for your help Vincent<br/> <br/> <cite>Vincent Granville said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/regression-analysis?commentId=6448529%3AComment%3A695879&xg_source=msg_com_forum#6448529Comment695879"><div><div class="xg_user_generated"><p>You could turn your categorical variables into dummy binary variables (google "dummy variable"), and then use any standard regression. This is often done in the context of logistic regression.</p>
</div>
</div>
</blockquote> You could turn your categoric…tag:www.datasciencecentral.com,2018-02-22:6448529:Comment:6958792018-02-22T00:08:57.424ZVincent Granvillehttps://www.datasciencecentral.com/profile/VincentGranville
<p>You could turn your categorical variables into dummy binary variables (google "dummy variable"), and then use any standard regression. This is often done in the context of logistic regression.</p>
<p>You could turn your categorical variables into dummy binary variables (google "dummy variable"), and then use any standard regression. This is often done in the context of logistic regression.</p>