NAME: Nikko P.

Emping DATE: April 30, 2024

OFFER CODE & SCHEDULE: 6497 (10:30 A.M – 12:00 P.M)

1. Differentiate fixed effect from random effect in Panel Data Model

Fixed Effect Random Effect
1. This approach treats This approach treats characteristics
characteristics or effects or effects specific to the individual or
specific to the individual or entity as random variables and are
entity as constant and will not drawn from a population
change overtime distribution.

2. Dummy variables are Believed to be unrelated to the

incorporated into the dataset explanatory variables that have
for every individual or entity to been observed. Rather, they are
absorb individual-specific represented as arbitrary departures
effects. from the population mean.

3. Enable the estimation of the Enable the estimation of variation

within-group variation, within and between groups,
accounting for any capturing effects that are distinctive
unobserved individual-specific to an individual over time that are
characteristics and capturing both time-invariant and time-
the effects of time-invariant varying.
variables.
4. Usually calculated using Estimated by utilizing data from both
within-group variation, which inside and across panel members or
limits the information used to entities, and combining within-
changes over time inside each group and between-group
individual or business. variation.

2. Differentiate Cross Sectional Data from Time Series Data

- Cross sectional data pertains to observations made at a single point in
time across various people, things, or groups make up cross-sectional
data. Furthermore, it records data on various topics or units at a
particular time so that they can be compared. Time Series Data on the
other hand is composed of observations gathered for a particular
person or unit over a series of time intervals. It permits the study of trends,
 patterns, and dynamics by following the behavior or evolution of a
variable or phenomenon across time.

3. When do you use Fixed Effect Data Model?

- In panel data analysis, fixed effects models are used to account for
unobserved, time-variable individual-specific characteristics.
Assume you have information about people or things that have been
observed throughout a number of time periods. These people or
things might possess special qualities or traits that aren't explicitly
quantified in your dataset, yet they might nonetheless have an
impact on the results you're after. You may effectively establish a
distinct category or "dummy variable" for every person or thing in your
dataset by utilizing a fixed effects model. This dummy variable
records all the unobserved attributes unique to that person or thing
that remain constant across time. Therefore, you are indirectly
accounting for these traits in your study rather than measuring them
openly.

4. When do you use Random Effect Panel Data Model?

- Random effects models interpret unobserved individual-specific
effects as random variables chosen from a population distribution, in
contrast to fixed effects models, which compensate for them by
considering them as constants. The random effects model excels in
a number of instances presented by this nuanced approach.
Random effects models are especially useful when the goal of the
study is to find average impacts or broad patterns among individuals
or entities in a community. Random effects models allow researchers
to capture the average impact of variables on the outcome of
interest, while noting that individual-specific effects may range
around this average, as opposed to isolating traits distinct to an
individual.

5. When do you use Dynamic Panel Data Model?

- Models of dynamic panel data are well-suited to investigate the
dynamics of social or economic processes as they develop over
time. In macroeconomics, for example, academics frequently
examine the dynamics of production growth, consumption, and
 investment, where variables may show both long- and short-term
trends. Insights into the development of economic phenomena can
be gained by examining lagged effects, feedback mechanisms,
and shock persistence using dynamic panel data models. For the
analysis of longitudinal data with temporal dynamics, persistence,
and dependency, dynamic panel data models are useful
instruments. Researchers can learn more about the dynamics of
social, economic, and spatial processes across time by using these
models, which incorporate lagged effects, account for
heterogeneous responses, and handle endogeneity problems.

6. Differentiate Auto-Correlation from Hetereroskedasticity

- In Auto-Correlation it refers to the correlation between a regression
model's error terms, or residuals, at various time intervals or
observations. It happens when the error terms show some degree of
association over time but are not independent from one another. The
premise of observational independence, which is necessary for the
validity of regression analysis, is broken by autocorrelation.
Heteroskedasticity on the other hand, is when the variance of the
error terms (residuals) in a regression model fluctuates at different
values of the independent variables. It happens when the
explanatory variable values cause a systematic shift in the residuals'
spread or dispersion. The assumption of constant variance of errors,
which is required to produce objective and effective estimates of
model parameters, is broken by Heteroskedasticity.

7. What is the role of Hausman test in choosing either Fixed Effects or

Random Effects Panel Models?
- The role of the Hausman test is to evaluate the consistency of Given
estimators under several hypotheses regarding effects that are
unique to each individual, this statistical test aids researchers in
choosing the best model for their investigation. The test evaluates
whether the discrepancies between the FE and RE estimators are
random or systematic by comparing them. Researchers should use
the FE model since it yields consistent estimates even in cases where
individual-specific effects are correlated with observed variables,
provided that the Hausman test rejects the null hypothesis,
suggesting systematic and substantial differences. On the other
 hand, due to its efficiency improvements in capturing average
effects among people or entities within the population, the RE model
can be favored if the test is unable to reject the null hypothesis,
indicating random and inconsequential differences.

8. Differentiate Balance from Unbalance Panel Data

Balanced Panel Data Unbalanced Panel Data
1. Includes observations for each Contain observations for individuals
person or thing in the panel for or entities that may vary across
each time span in the dataset. different time periods in the dataset.

2. Over the whole time period Due to non-response, attrition, or

covered by the data, every other factors, some people may
person or entity is seen have missing observations for
consistently—there are never specific time periods; this could lead
any missing observations. to inconsistent or missing data across
the panel.
3. This consistency in Because of the non-uniformity in the
observations across all subjects observations, analysis is made more
and time periods makes difficult because it necessitates
analysis easier and permits the taking into account missing data
use of some statistical and maybe modifying estimating
approaches that call for approaches to account for panel
complete data for every unit structure variability.
at every time point.

9. Differentiate Static Panel Data Analysis from Dynamic Panel Data Analysis
- Static panel data analysis focuses on analyzing correlations between
variables using data collected from several individuals or entities
observed during a given period of time without taking into account
the temporal dynamics or interdependence of observations.
Estimating the average impact of explanatory variables on the
outcome of interest across all persons or entities in the panel is the
main focus of static panel data analysis. This method ignores lagged
effects, feedback mechanisms, and shock persistence in favor of the
constant relationships between variables over time.
- On the other hand, simulating the temporal dynamics, persistence,
and interdependence of observations throughout time is the main
goal of dynamic panel data analysis. In contrast to static panel data
analysis, dynamic panel data analysis recognizes that lagged
effects, feedback mechanisms, and shock persistence all play
important roles in how the relationships between variables may
change over time. By modeling how variables change and interact
over time, researchers may capture both short-term fluctuations and
long-term trends in their analysis of dynamic panel data.