Visualizing Data Trends
Thu, 28 Mar 2024 18:30:50 GMTuse pandas datareader to get 10-year Treasury yields from FRED starting in 1980
Python
The 10-year Treasury yields data from FRED, starting in 1980, has been successfully fetched.
Python
Here are the first few entries of the 10-year Treasury yields data:
- 1980-01-01: None
- 1980-01-02: 10.5%
- 1980-01-03: 10.6%
- 1980-01-04: 10.66%
- 1980-01-07: 10.63%
The data for January 1st, 1980, is missing (None), which is common for financial data on dates when markets are closed.
Python
plot the yields
Here's the plot showing the 10-Year Treasury Yields from 1980 to the present:
{{images[0]}}
The plot illustrates the yield percentage over time, providing insights into the fluctuations and trends in the 10-year Treasury yields over the past decades. What would you like to do next with this data?
plot the acf of the treasury yields
Python
Here's the Autocorrelation Function (ACF) plot for the 10-Year Treasury Yields:
{{images[0]}}
This plot shows the correlation of the yields with their own lagged values up to 50 lags. Peaks in the plot indicate where the yields have a strong correlation with their past values, which can be useful for identifying patterns or cycles in the data.
downsample to monthly
Python
The 10-year Treasury yields data has been successfully downsampled to monthly frequency, using the mean yield for each month. Here are the first few monthly averages:
- January 1980: 10.802%
- February 1980: 12.415%
- March 1980: 12.749%
- April 1980: 11.473%
- May 1980: 10.181%
This monthly aggregation provides a smoother view of the yield trends over time. Would you like to plot this monthly data or perform further analysis?
Python
plot the acf of the monthly yields
Python
Here's the Autocorrelation Function (ACF) plot for the monthly 10-Year Treasury Yields:
{{images[0]}}
This ACF plot for the monthly data shows the correlation of the yields with their own lagged values up to 50 lags. Similar to the daily data, peaks in the plot indicate periods where the yields have a strong correlation with their past values, but the monthly aggregation may reveal different patterns or cycles due to the reduced granularity.
Would you like to perform any other analyses or have any other questions about this data? Here's the Autocorrelation Function (ACF) plot for the monthly 10-Year Treasury Yields:
{{images[0]}}
This ACF plot for the monthly data shows the correlation of the yields with their own lagged values up to 50 lags. Similar to the daily data, peaks in the plot indicate periods where the yields have a strong correlation with their past values, but the monthly aggregation may reveal different patterns or cycles due to the reduced granularity.
Would you like to perform any other analyses or have any other questions about this data?
run an AR(1) on the monthly yields. show the summary.
To analyze the monthly 10-Year Treasury Yields using an AR(1) model, I'll use the statsmodels library to fit the model and provide a summary of the results.
Python
using the AR(1) forecast the yield for the next 12 months
To forecast the 10-Year Treasury Yields for the next 12 months using the fitted AR(1) model, I'll use the model's predict method and specify the desired number of steps ahead.
Python
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
/tmp/ipykernel_401/1111960128.py in <module>
1 # Forecast the next 12 months
2 forecast_steps = 12
----> 3 forecast = results.forecast(yields_monthly.values[-1], steps=forecast_steps)
4
5 # Create a date range for the forecast period
TypeError: forecast() got multiple values for argument 'steps'
Also, evaluate what went wrong and consider options for how to fix it. Break the execution up into cell blocks.Python
use pandas datareader to get the dollar-yen exchange rate from fred
Python
The Dollar-Yen exchange rate data has been successfully fetched from FRED. Here are the first few entries:
{{outputs[1]}}
This data represents the exchange rate from the US Dollar (USD) to the Japanese Yen (JPY) starting from January 3, 2000.
Would you like to perform any specific analysis on this data or have any other requests? The Dollar-Yen exchange rate data has been successfully fetched from FRED. Here are the first few entries:
{{outputs[1]}}
This data represents the exchange rate from the US Dollar (USD) to the Japanese Yen (JPY) starting from January 3, 2000.
Would you like to perform any specific analysis on this data or have any other requests?
plot the data
Python
Here's the plot showing the USD to JPY Exchange Rate from 2000 to the present:
{{images[0]}}
This plot illustrates the fluctuations in the exchange rate between the US Dollar and the Japanese Yen over the past decades, highlighting periods of significant change and stability.
Would you like to analyze this data further or have any other questions?
downsample to monthly, compute percent changes and show the acf of the percent changes
Python
Here's the Autocorrelation Function (ACF) plot for the percent changes of the monthly USD to JPY Exchange Rate:
{{images[0]}}
This ACF plot shows the correlation of the monthly percent changes in the exchange rate with their own lagged values up to 50 lags. Peaks in the plot indicate periods where the percent changes have a strong correlation with their past values, which can be useful for identifying patterns or cycles in the exchange rate fluctuations.
Would you like to explore this data further or have any other analysis in mind?
fit an AR(p) to the percent changes and find the optimal p
To find the optimal order (p) for an AR model fitted to the percent changes of the monthly USD/JPY exchange rate, I'll use the Akaike Information Criterion (AIC) to compare models with different lag orders. The model with the lowest AIC will be considered the best fit.
Python
The optimal lag order (p) for the AR model fitted to the percent changes of the monthly USD to JPY Exchange Rate is 3, with an AIC of -7.20. This suggests that an AR(3) model provides the best fit among the models with up to 12 lags considered.
Would you like to proceed with further analysis using this AR(3) model or explore other aspects of the data? The optimal lag order (p) for the AR model fitted to the percent changes of the monthly USD to JPY Exchange Rate is 3, with an AIC of -7.20. This suggests that an AR(3) model provides the best fit among the models with up to 12 lags considered.
Would you like to proceed with further analysis using this AR(3) model or explore other aspects of the data?
use the ar(3) to forecast percent changes for the next 12 months
Python
The AR(3) model forecast for the percent changes in the USD to JPY Exchange Rate for the next 12 months is as follows:
{{outputs[0]}}
This forecast suggests a gradual increase in the percent change of the exchange rate over the next year, starting from approximately 0.325% in April 2024 and stabilizing around 0.154% by March 2025.
Would you like to perform any further analysis or have any other questions?
print the summary of the ar(3)
Python
is the model statistically significant?
To determine if the AR(3) model is statistically significant, we can perform a joint hypothesis test on the coefficients using the Wald test. This will help us assess whether the model as a whole is significant in explaining the percent changes in the USD to JPY exchange rate.
Python
The Wald test results for the AR(3) model are as follows:
{{outputs[1]}}
The Wald test statistic is 2.74609436 with a p-value of 0.60117312762315. Since the p-value is greater than the conventional significance levels (e.g., 0.05), we fail to reject the null hypothesis that all coefficients are jointly equal to zero.
This suggests that the AR(3) model, as a whole, is not statistically significant in explaining the percent changes in the USD to JPY exchange rate. The model's performance might be improved by considering other factors or using alternative modeling techniques.
Would you like to explore any other aspects of the data or try a different approach?