This dataset contains housing-related information, including: - Geographic features (longitude, latitude) - Housing characteristics (housing_median_age, total_rooms, total_bedrooms, households) - Socioeconomic indicators (median_income, median_house_value) - Location (ocean_proximity)

This Kaggle dataset comes from the 1990 California census. While outdated for predicting current housing prices, it provides valuable insights into historical trends and serves as a useful dataset for exploring fundamental data analysis techniques.

1. What factors contribute most to housing prices?
2. Is there a geographic trend in house prices?
3. How does population density relate to house prices?
4. Does ocean proximity influence housing prices?
5. Can non-linear models provide better predictions?

This histogram shows the distribution of median house values in the dataset. The data is right-skewed, with most houses priced under $500,000, however, some high-value homes contribute to a long tail.

This scatter plot shows a strong positive correlation between median income and house value. Higher-income neighborhoods tend to have higher house prices, as expected.

states <- ne_states(country = "united states of america", returnclass = "sf")
california <- states[states$name == "California", ]

ggplot() +
  geom_sf(data = california, fill = "gray90", color = "black") +
  geom_point(data=data, aes(x=longitude, y=latitude, color=median_house_value), alpha=0.5) +
  scale_color_gradient(low="blue", high="red", labels = scales::dollar_format()) +
  ggtitle("Geographic Distribution of House Prices Over California") +
  xlab("Longitude") +
  ylab("Latitude")

Here, we see that coastal cities tend to have higher home prices (red). Inland areas tend to have lower-priced homes, likely due to reduced demand and different economic opportunities.Additionally, coastal cities tend to be more desirable due to their proximity to the ocean, moderate climate, and access to amenities such as beaches, tourism, and commercial centers.

This plot reveals a negative exponential decay relationship—as population density increases, house values decrease at a declining rate. Crowded areas often have smaller homes due to space constraints, thus lower prices.

\[ Y = a e^{-bX} \] where: - \(Y\) is the median house value - \(X\) is the population density - \(a\) represents the initial house value when density is low - \(b\) controls the rate of decay This equation models the sharp decrease in house values as population density increases, following an exponential decay pattern.

Houses near the ocean tend to be significantly more expensive as compared to houses inland.

This 3D scatter plot visualizes the relationship between median income, house value, and house age. We can observe patterns in pricing based on economic factors and home age, offering a more detailed view of the data.

• Income is a strong predictor of house value.
• Coastal areas tend to have higher home prices.
• Population density follows an exponential decay relationship with house prices.
• Ocean proximity significantly affects housing prices.