Question

In: Economics

1. Suppose you run a regression on past data to find the magnitude of consumption responses...

1. Suppose you run a regression on past data to find the magnitude of consumption responses to changes in interest rates. What is the main error that you are making if you simply use this number to determine the optimal interest rate change to respond to an economic shock?

2. Why can limiting the choices of policy makers result in better long-term outcomes?

Solutions

Expert Solution

Taking my time to write answers worth the time. Please leave an upvote ?, I appreciate your support :)

-----

1) The main mistake is assuming: ceterus paribus.

Let me explain. In cybernetics there's the concept of a black box. Let's take the case of an aeroplane for example. You hit the button, the aeroplance takes off. From this, an amateur at observational inference would draw a causal relationship between the pressing of the button and the aeroplance taking off - it is the pushing of the button which causes the aeroplane to go into flight. This is because the inner workings of the plance are invisible to us - it is a black box.

In reality, the button does not cause: it only triggers a chain of events to start inside the aeroplance body which eventually culminates in the aeroplance taking off.

The difference is that in the absence of an intermediate chain i.e. a direct causal relation, if the antecedent condition is satisfied then the subsequent condition must be executed.

Whereas for a chain, the forward transmission of the initiating signal may break somewhere along the chain - which would prevent the final condition (aeroplane takes flight) from executing.

Other factors have a role to play at different parts of the chain - they do vary across time and circumstance, so ceterus paribus is unrealistic.

Similarly in an economic system, the change in interest rate is the initiating signal. The intermediate chain of effect is through Investment then Output then National Income then Disposable Income and then Consumption.

Think of the intermediate points on the chain as switches: the initiating signal does or does not go forward to finality depending upon which positions the switches are in. There can be a number of unpredictable influences (shocks) on these switches such that the initiating signal (change in interest rate) fails to travel all the way through to bring about the desired result.

----

But another important part of this answer should be that a proper economic shock may be a Black Swan event - unprecedented in qualitative and quantitative features, so cannot be tackled on models driven by past data alone.

During an economic shock, many implicit assumptions about the nature of the social fabric aren't valid anymore. These assumptions are so implicit, many if not most wouldn't even recognize them as assumptions since they're so taken for granted.

It's only when the contrary has been observed i.e. contextual non-validity of such an assumption, does it come under attention as an assumption - as an extrinsic (subjective, imposed by observer) rather than an intrinsic feature of the system under observation.

When these assumptions don't hold, "Economics" doesn't hold. Axioms are like pillars for a theoretical model. If the pillars aren't there, the model falls.

-----

2) This is the idea of via negativa, popularized by statistician and thinker Nassim Nicholas Taleb - it can be translated as "by the negative" or "doing by non-doing", i.e. shifting focus from what to do to what not to do.

It finds echo at least in Bruce Lee (whose martial art Jeet Kune Do translates roughly to "The Way of No Way")

and Far Eastern philosophy in general (Tao te Ching: "A truly good man does nothing, Yet nothing is left undone. A foolish man is always doing, Yet much remains to be done.")

In the modern context, policy makers are subject to the human frailty of informational inadequacy and emotional impulsivity. This can cause them to "overreact" to natural fluctuations in economic variables, microadjusting and overoptimizing at every juncture. These microadjustments need further adjustments, and the computational complexity of the task scales too rapidly for us (humans) to keep up.

Soon a narrative of resultant (unexpected) glitches and patchwork fixes is being written. Better that policy intervention be limited in number to the few cases where they are unambiguously warranted.


Related Solutions

After you run the Regression, answer these questions (type responses in this box): 1. What is...
After you run the Regression, answer these questions (type responses in this box): 1. What is the p-value (the significance)? 2. Is the relationship between Hours in Therapy and Depression Inventory Score statistically significant? 3. Do Depressions Scores increase or decrease as a result of having more Hours in Therapy? REGRESSION Hours in Therapy Depression Inventory Score 9 20 9 28 9 20 9 26 11 19 11 15 13 12 13 11 15 10 15 11 16 10 16...
Question 1. In this exercise you will simulate a data set and run a simple regression....
Question 1. In this exercise you will simulate a data set and run a simple regression. To ensure reproducible results, make sure you use set.seed(1). a)   Using the rnorm() function create vector X that contains 200 observations from N(0,1) distribution. Similarly, create a 200 element vector, epsilon (ϵ), drawn from N(0,0.25) distribution. This is the irreducible error. b)   Create the response data using the following relationship: Y=−1+0.5X+ϵ Fit a linear regression of Y on X. Display the summary statistics and...
You run a regression analysis on a bivariate set of data (n=99). You obtain the regression...
You run a regression analysis on a bivariate set of data (n=99). You obtain the regression equation y=0.843x+6.762 with a correlation coefficient of r=0.954 which is significant at α=0.01 You want to predict what value (on average) for the explanatory variable will give you a value of 150 on the response variable. What is the predicted explanatory value? x = (Report answer accurate to one decimal place.) Here is a bivariate data set. Find the regression equation for the response...
Run two regressions using Excel from the data below. Find the following information: 1. estimated regression...
Run two regressions using Excel from the data below. Find the following information: 1. estimated regression equations for both regressions 2. both coefficients of determination. 3. significance of each independent variable 4.Report the significance of both models. 5.Predict y for a fictitious set of x values for both Years Weekend Daily Tour Income Daily Gross Revenue Number of Tourists 1 Friday 3378 4838.95 432 1 Saturday 1198 3487.78 139 1 Sunday 3630 4371.3 467 2 Friday 4550 6486.48 546 2...
(3) Suppose that you run a regression of wage on a set of independent variables: age,...
(3) Suppose that you run a regression of wage on a set of independent variables: age, years of education, experience, and a dummy variable married. There are 100 observations, i.e., n=100. You want to test whether all the coefficient and intercept estimates in the equation for married and unmarried are equal to each other. i.e., Married: wage = β0 + β1*age + β2*educ + β3*exper + uUnmarried: wage = α0 + α1*age + α2*educ + α3*exper + v (a) Write...
1. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this regression using OLS and...
1. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this regression using OLS and get the following results: b0=-3.13437; SE(b0)=0.959254; b1=1.46693; SE(b1)=21.0213; R-squared=0.130357; and SER=8.769363. Note that b0 and b1 the OLS estimate of b0 and b1, respectively. The total number of observations is 2950.According to these results the relationship between C and Y is: A. no relationship B. impossible to tell C. positive D. negative 2. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this...
1. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this regression using OLS and...
1. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this regression using OLS and get the following results: b0=-3.13437; SE(b0)=0.959254; b1=1.46693; SE(b1)=0.0697828; R-squared=0.130357; and SER=8.769363. Note that b0 and b1 the OLS estimate of b0 and b1, respectively. The total number of observations is 2950. The number of degrees of freedom for this regression is A. 2952 B. 2948 C. 2 D. 2950 2. Consider the model Ci= β0+β1 Yi+ ui. Suppose you run this regression using OLS...
Suppose you run a regression containing observations for each of the 74 kinds of cars released...
Suppose you run a regression containing observations for each of the 74 kinds of cars released in 1978 in the United States, and you regress price (in dollars) on weight (in pounds). You get the following results: ˆ β0 is -6.71, with SE 1174.4. ˆ β1 (slope coefficient on weight) is 2.04 with SE .377. • (a) Say, in words, what the slope coefficient means in this case, without taking a stand on causality. • (b) Suppose I give you...
Run a linear regression using Excel’s Data Analysis regression tool. Construct the linear regression equation and...
Run a linear regression using Excel’s Data Analysis regression tool. Construct the linear regression equation and determine the predicted total sales value if the number of promotions is 6. Is there a significant relationship? Clearly explain your reasoning using the regression results. Number of Promotions Total Sales 3 2554 2 1746 11 2755 14 1935 15 2461 4 2727 5 2231 14 2791 12 2557 4 1897 2 2022 7 2673 11 2947 11 1573 14 2980
A) Running a regression on 11 pairs of data, you find the correlation coefficient to be...
A) Running a regression on 11 pairs of data, you find the correlation coefficient to be 0.321. At the .05 significance level, does this show significant correlation? Yes, significant correlation No B) At a .01 significance level with a sample size of 50, find the critical value for the correlation coefficient
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT