Scribd Logo
Upload

Welcome to Scribd!

  • 24 views

    Samsung Quarterly Sales Prediction Report

    Uploaded by

    prathmeshsuryavanshi97
    1) The report aims to predict Samsung's quarterly sales using a time series analysis of 12 years of past data. Initially, a linear model was used but did not accurately capture the trend. 2) A quadratic model was then applied, squaring the time variable, which better fit the data. This quadratic model successfully predicted Samsung's 2023 quarterly sales values. 3) An iterative approach of testing different models was important to identify the best representation of Samsung's sales trajectory over time. The quadratic model proved to most accurately forecast quarterly sales.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Samsung Quarterly Sales Prediction Report

    Uploaded by

    prathmeshsuryavanshi97
    24 views4 pages
    1) The report aims to predict Samsung's quarterly sales using a time series analysis of 12 years of past data. Initially, a linear model was used but did not accurately capture the trend. 2) A quadratic model was then applied, squaring the time variable, which better fit the data. This quadratic model successfully predicted Samsung's 2023 quarterly sales values. 3) An iterative approach of testing different models was important to identify the best representation of Samsung's sales trajectory over time. The quadratic model proved to most accurately forecast quarterly sales.

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1) The report aims to predict Samsung's quarterly sales using a time series analysis of 12 years of past data. Initially, a linear model was used but did not accurately capture the trend. 2) A quadratic model was then applied, squaring the time variable, which better fit the data. This quadratic model successfully predicted Samsung's 2023 quarterly sales values. 3) An iterative approach of testing different models was important to identify the best representation of Samsung's sales trajectory over time. The quadratic model proved to most accurately forecast quarterly sales.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    24 views4 pages

    Samsung Quarterly Sales Prediction Report

    Uploaded by

    prathmeshsuryavanshi97
    1) The report aims to predict Samsung's quarterly sales using a time series analysis of 12 years of past data. Initially, a linear model was used but did not accurately capture the trend. 2) A quadratic model was then applied, squaring the time variable, which better fit the data. This quadratic model successfully predicted Samsung's 2023 quarterly sales values. 3) An iterative approach of testing different models was important to identify the best representation of Samsung's sales trajectory over time. The quadratic model proved to most accurately forecast quarterly sales.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 4

    Samsung Quarterly Sales Prediction Report

    Subject Name
    Managerial Economics

    Submitted to
    Dr. Manish Sinha

    Submitted by
    Prathmesh Suryavanshi (23020348029)
    Introduction

    The objective of this report is to predict Samsung's quarterly sales.


    Initially, we attempted to forecast sales using a linear model, taking into
    account the seasonality of the data. However, upon analysis, it was found that
    this model was not accurate enough. Therefore, we pivoted to a quadratic
    model, squaring the time variable, which yielded improved results. In our
    scenario, we have taken Samsung 12 years of quarterly data (2010 to 2023).
    We will try to predict the sales of 2023 by using the optimal trend analysis
    method.

    Methodology

     Removing Seasonality
    To ensure a fair comparison between actual sales and our predicted
    values, the seasonality inherent in Samsung's sales data was first removed.
    This step was crucial to isolate the trend component of the data and make
    accurate predictions.

     Linear Prediction Model


    Once seasonality was removed, we applied a linear regression model to
    predict quarterly sales. This approach was chosen based on the initial
    observation of the trend, which appeared linear.

    Prediction
    100
    90
    80
    70
    60
    50
    40
    30
    20
    10
    0
    0 10 20 30 40 50 60
     Quadratic Prediction Model
    Given that our linear model did not deliver accurate predictions, we
    revisited our approach. It was hypothesized that the trend may follow a
    quadratic curve instead. By squaring the time variable, we transformed our
    model from a linear to a quadratic one, expecting a better fit. We can see
    direction in the data only once.

    Chart Title
    100
    90
    80
    70
    60
    50
    40
    30
    20
    10
    0
    0 500 1000 1500 2000 2500 3000 3500

    Results

     Linear Prediction Model Outcomes


    The linear model showed significant discrepancies between the
    predicted and actual quarterly sales values. This indicated that a linear trend
    might not be the best representation of Samsung's sales trajectory over the
    given period.

    Actual Sale Sales


    Predicted
    535 Q1'23 60.50
    85.3217746
    543 Q2'23 53.50
    555 Q3'23
    81.6412932
    564 Q4'23
    5
    90.2345144
     Quadratic Prediction 5 Model Outcomes
    5
    87.4469434
    6
    The quadratic model, on the other hand, matched the actual trend. This
    alignment confirmed our hypothesis that the sales trend had a quadratic
    nature.

    Actual Sale Predicted Sales


    53 Q1'23 60.50 5 54.4626803
    54 Q2'23 53.50 3 9
    55 Q3'23 5
    53.7892671
    56 Q4'23 4
    5 45.6372349
    Discussion 5 2
    5 43.7985032
    The initial 6 2 understanding of the sales
    trend highlights the importance of model selection in time series forecasting.
    While linear trends are common and can be an appropriate starting point, it's
    crucial to verify the model's accuracy with the actual data. In this case, our
    iterative approach allowed us to refine our prediction model, we tired the
    cubic quadratic model and exponential model and ultimately selecting the
    square quadratic model as the best representation of Samsung's sales trend.

    Conclusion

    Through the process of removing seasonality, testing a linear model, and


    then pivoting to a quadratic model, we were able to accurately predict
    Samsung's quarterly sales. This aspire underscores the importance of flexibility
    and adaptability in data analysis, especially when predicting complex real-
    world phenomena like sales trends.

    You might also like