Want to create interactive content? It’s easy in Genially!

Get started free

Alpek

Manu Reyna

Created on March 26, 2024

Start designing with a free template

Discover more than 1500 professional designs like these:

HR Organizational Chart

SWOT PRO

Branching diagram

Fishbone Diagram

Puzzle Diagram

Gear Diagram

Square Timeline Diagram

Transcript

ALPEK

Made by:

April 5th, 2024

Moving Average

EWMA

What is Alpek?

This model gives more weight to recent data and less weight to older data. It uses the method of maximum likelihood to determine the best value for lambda, ensuring accurate alignment with the data.

This method calculates variance for a day "t" using data from the preceding "m" days. It assigns equal significance to all returns within this period when computing the variance.

Founded in 1994, Alpek stands as a prominent Mexican company in the petrochemical and textile sectors, renowned for its global presence and contributions. At its core, Alpek is recognized for its leadership in producing PTA, PET, EPS worldwide, and for being a top rPET manufacturer in the Americas.

Characteristics

Results

Characteristics

Graphs

Errors

Results

Lambda

Locations

Price

ARCH

GARCH

The GARCH model extends ARCH with added flexibility, addressing changing volatility over time. It's a robust framework that models variance as a function of past squared observations and forecast errors.

This model express the variance as a function of past squared observations. It acknowledges that volatility is not constant but instead varies over time. Essentially, ARCH tells us that future volatility depends on past volatility.

EPS

Employees

Parameters

Constant

Arch Test

Non Constant

Characteristics

Recap

Differences

Returns

M:
  • 5
  • 10
  • 20
  • 40

Test

What does this test refer to?

2×10

-16

The ARCH-GARCH effects test assesses whether there's conditional heteroskedasticity in a dataset, meaning if there are persistent volatility patterns in a model's squared errors, indicating ARCH-GARCH effects.

  • Null Hypothesis (Ho): There are no ARCH-GARCH effects.
  • Alternative Hypothesis (Ha): There are ARCH-GARCH effects.

p value
graph

Estimation

After analysing results we end up in conclusion that the optimal value of lambda is 0.969. This gives us the volatility for tomorrow of:

11.62%

Constant

Estimation
Results

Estimation

After analysing results we end up in conclusion that the optimal value of m is 5. This gives us the volatility for tomorrow of:

1.732%

Formula

Sum of the squared returns for the past m days

Variance at 't' days and 'm' meaning the # of data used

Squared returns from pevious days

Weights

Formula

Gives more importance to recent returns

Gives more importance to recent variance

Variance at 't' days

Squared returns from pevious days

Variance from pevious days

Allows future estimations

Parameters

We need to ensure our model optimizes this three parameters:
  • AIC: Measures model quality relative to the amount of information used to estimate it.
  • BIC: Similar to AIC but penalizes model complexity more strongly.
  • Likelihood Function: Represents how likely the estimated parameters of the model produced the observed data.

Formula

Gives more importance to recent returns

Variance at 't' days

Squared returns from pevious days

Allows future estimations

Non Constant

We have a time series given by:

Arima

Results
Locations

With 33 locations worldwide, Alpek operates manufacturing and processing facilities across the Americas, the United Kingdom, and the Middle East, employing over 7,000 individuals globally. By strategically placing our state-of-the-art facilities, we've generated opportunities for cost-effective logistics, synergizing our diverse operations in a unique and optimized manner.

Plastics and Chemicals Polyester

Moving Average

Characteristics
  • Easy to understand and implement.
  • Smooths out fluctuations in data over time.
  • It's a lagging indicator, meaning it reacts to past data rather than predicting future trends.
  • Assigns equal weight to all data points within the specified period.
  • Can be sensitive to outliers, especially when using short time periods.
  • It's best suited for short-term forecasting due to its reliance on historical data.
Historic Price

ALPEK's price history has seen significant fluctuations, notably in the third quarter of 2023 when a 10% drop occurred due to disappointing quarterly results. Revenues fell by 33.7%, attributed to factors like normalized freight prices and market saturation. Sales in polyester and plastics/chemicals divisions notably declined by 38.7%.

Errors

Formulas
Characteristics

EWMA

Characteristics
  • Past squared returns' influence decreases exponentially over time.
  • It adjusts to changes in data patterns because it gives higher weight to recent observations.
  • Provides a smoother estimation of variance compared to Moving Average.
  • Can be customized by adjusting the smoothing parameter to match the data's volatility pattern.
  • Requires fewer computational resources compared to other forecasting methods due to its simple calculation process.
Earnings per share

Over the past two years, Alpek has demonstrated ups & downs in its Earnings Per Share (EPS), reflecting the company's volatility. Despite challenges in the global market, Alpek's EPS has shown resilience, trying to keep a steady growth.

Lambda and Maximum Likelihood

Formula

Squared returns from yesterday

Variance at 't' days

Last day variance estimate

The rate at which the impact of previous squared returns diminishes over time in an exponential manner.