Error model is singular

I have the following data set:

                 |       Scenario 1       |     Scenario 2         |
                 |Trial 1|Trial 2| Trial 3|Trial 1|Trial 2| Trial 3|
 -------------------------------------------------------------------
              S1 | ...
 Condition 1  S2 | ...
              S3 |
 -------------------------------------------------------------------
              S5 |
 Condition 2  S6 |
              S7 |

Thus the Trials are nested in the Scenarios and all of them are within subject. I am trying to run an ANOVA on this data set. Here is the model without defining that Scenarios (and Trials) are within subject.

 my_data.aov <- aov(value~Condition*Trial%in%Scenario,data=my_data) #works fine

But when I specify that these are within subject:

my_data.aov <- aov(value~Condition*Trial%in%Scenario+Error(Player/(Trial%in%Scenario)),data=my_data) 

I get the following error

In aov(value ~ Condition * Trial % in % Scenario + Error(Player/(Trial %in%  :
Error() model is singular

The closest set-up I could find was Split plot in R but there the subjects are nested inside each Trial not in each Condition.

EXAMPLE FILE

Here is an example file in long format.

What about this approach?

If I treat each Trial as a sample, then I can collapse across Scenarios by averaging them, so I will have a simpler model, where each Subject‘s behavior is described per Scenario. And since I need to analyze the relationship of value~Condition*Scenario I can do so by defining the Error like Error(Subject/Scenario).

Will this approach invalidate my analysis?

I am trying to run an ANOVA on a 5×5 design, where each subject only saw 5 of the 25 possible conditions. So subject A saw,

1,1; 2,2; 3,3; 4,4; 5,5 

and subject B saw

1,2; 2,3; 3;4, 4;5, 5,1 etc.

Is it incorrect to use a within subject ANOVA since its not a true within subjects design?

This is what I was using,

(aov(Risk ~ Moral * Age + Error(ID/(Moral * Age)), data = Home)

and I get

Warning in aov(Risk ~ Moral * Age + Error(ID/(Moral * Age)), data = Home) :
  Error() model is singular

Here is a sample of the data

          ID   Age  Moral Risk
1         90   two affair    2
2         90 eight    vol    2    
3         90   ten  invol    2    
4         90  four   work    3   
5         90   six    gym    4    
6        117   two affair    9    
7        117 eight    vol    2    
8        117   ten  invol    3   
9        117  four   work    9  
10       117   six    gym   10  
11        70   two affair    3 
12        70 eight    vol    3 
13        70   ten  invol    2
14        70  four   work    6
15        70   six    gym    6
16        51   two affair    4
17        51 eight    vol    3
18        51   ten  invol    3
19        51  four   work    4
20        51   six    gym    6

Note: See
TracTickets for help on using
tickets.

Random effects model with PLM: «System is computationally singular»-Error?

I am currently trying to estimate some panel data models in R using PLM package. This includes the estimation of basic pooled, fixed effects and random effects models. Therefore I make use of this code:

Now here’s the problem: I can without any problem estimate all models except for the random effects model. After entering the «random»-formula, R produces the following error:

• Linear combinations in x? A first guess would be that there are exact linear dependencies of the exogenous variables in x. The data is balance sheet data and I would like to explain the standard deviation (y) of a specific balance sheet position by other balance sheet positions (or the ratio of the position and the balance sheet sum). Of course, the variables in x are related to each other. For example some of the ratios are calculated by dividing by the mean which is also a separate independant variable. And the dependant variable, which is the standard deviation, is also calculated by using this mean. But again: There should be no EXACT correlation. But: If I exclude some of my exogenous variables, the problem disappears, but I have to include them actually.
• Unbalanced panel / NAs? The data is unbalanced and there are NAs. Fixed effects output says: n=16, T=18-40, N=455. Probably the unbalanced data or the NAs are the reason for the error?

Is there anybody who can give me a hint what this error does actually mean and especially: how to solve the problem? How do I have to correct the code in order to get results? Thanks a lot!

Источник

VAR MODEL: Error in solve.default(Sigma) : system is computationally singular: reciprocal condition number

I am using R vars package to implement VAR model in a multivariate time series model. I tried to run:

but I was getting this error message:

Error in solve.default(Sigma) : system is computationally singular: reciprocal condition number = 0

foo_ts is a time series data.

I have tried to adapt examples in stackexchange and stackoverflow but have not been successful in running the VAR model. I have successfully run the VAR model on other datasets but not on foo_ts. Any help will be appreciated.

1 Answer 1

Your time series is too short to fit that many parameters. Recall that, in a VAR(p) model begin x_t = d + A_1 x_ + A_2 x_ + dots + A_p x_ + epsilon_t, end each parameter matrix $A_i$ will have dimension $ktimes k$ , where $k$ is the number of parameters. Estimation simply procedes by OLS to each equation of the VAR model. That means you will have $1+ktimes p$ parameters per equation (assuming $d$ for the deterministic component is just a constant). In your case, $p=6$ and there are $k=8$ variables (although two of these appear to be identically zero and hence not very useful).

Hence, you fit 49 coefficients for each equation, but only appear to have $n=21$ observations. As is well-known, the $X’X$ matrix of the OLS estimator will not be invertible when $n$ is smaller than the number of regressors.

Источник

Error in solve.default(VC) : system is computationally singular: reciprocal condition number = 3.40196e-17 #15

Dear,
Thank you again for this software. Unfortunately I am having trouble getting it to run on my own data. When I run the code below , I get the following error message from JointModel(): Error in solve.default(VC) : system is computationally singular: reciprocal condition number = 3.40196e-17 , Error: cannot allocate vector of size 86.8 Gb,
Error in lme.formula(obs

IND_SEVVP0 + AUDI0_C + SEXE + CENTRE + DIPNIV0C + :
nlminb issue, code d’convergence error = 1 message = false convergence (8)
Error in optim(thetas, opt.survWB, gr.survWB, method = «BFGS», control = list(maxit = if (it

The text was updated successfully, but these errors were encountered:

I am not sure to help because you got several errors but for the following error message

• Error in solve.default(VC) : system is computationally singular: reciprocal condition number = 3.40196e-17

I fixed it by scaling the values of the outcome with a standard transformation

N(0, 1)
You should also check if you have non-finite values ( Inf ) in your dataset and remove it

N(0, 1) You should also check if you have missing values or non-finite values (Inf) in your dataset and remove it — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub , or unsubscribe .

Источник

Getting a «system is computationally singular» error in sleuth

I am analysing 142 samples belonging to 6 batches. Additionally, those samples belong to 72 strains, which means that for most of the strains there are two samples.

I could fit simple models (for strain and batches for instance), but when I get to the «full» model (

batch+strain), I get the following error:

I should point out that of the 72 strains, only 15 have samples in distinct batches. This means that most strains (57) have both samples in the same batch.

Is the error due to an unknown bug or rather to the experimental design? Does it mean that the information on batches cannot be used?

EDIT I’ve posted the experimental design in a gist

2 Answers 2

You should be able to remove any one of the following strains to end up with a rank-sufficient model matrix: 5, 10, 12, 13, 14, 15, 19, 26, 28, 3, 30, 32, 36, 39, 41, 45, 46, 49, 5, 50, 52, 53, 58, 59, 60, 69, 8. As an aside you can figure this sort of thing out as follows (I read your dataframe in a the d object):

You obviously don’t want to remove the batch columns or the intercept. The normal tricks that you can sometimes use to get around this issue with case-control studies don’t appear to help here, which is why I would just drop a strain and call it done. Keep in mind that your power is still likely terrible. I generally recommend at least 6 replicates per group (scale down the number of groups to fit your budget).

Once the desired strain is removed from the model, it can be fit directly into the sleuth_fit function to obtain the full model:

Источник

Problem Description

What does the error message «Singular Matrix» mean?

Solution

The problem is that the stiffness matrix of the linear system is singular and the linear solver cannot invert it. Examples of practical modeling situations where this can occur are:

• One, or more, of the relevant material properties is zero. Or, the material properties become zero during the solution while solving a nonlinear problem. If you are solving a nonlinear problem, see also: Improving Convergence of Nonlinear Stationary Models. If you are solving a linear problem, see also: What to do when a linear stationary model is not solving

• Your problem (equation system combined with the boundary conditions) is over- or underspecified. Check the equations and boundary conditions. You will often see this error if you have an ill-posed initial condition. It can be a good start to look at the initial condition and set it to some educated guess. Another example can be if you apply periodic boundary conditions when the PDE problem you solve does not permit such a solution. Yet another example is a Navier-Stokes problem where the pressure is unconstrained on the boundaries. You will need to make a point constraint somewhere.

• You have specified a dependency of an expression defined in terms of coordinate derivatives of the solution variable. As the default initial guess into nonlinear systems is a constant (making the initial guess for the solution-derivative dependent expression zero), this can cause the equation to become singular.  The cure is to specify an initial value with a non-zero derivative, such as 1e-6*sqrt(x^2+y^2+z^2). This problem can sometimes occur when working with nonlinear materials, for example nonlinear magnetic materials (magnetic saturation effects).

• You are solving a nonlinear eigenvalue problem with a zero linearization point. Examples of models that often result in a nonlinear eigenvalue problem are eigenfrequency analyses with PMLs or lossy materials, in RF or acoustics. To avoid the error message, you need to specify a non-zero, possibly complex-valued eigenvalue linearization point in the Study->Solver Configurations->Solver->Eigenvalue Solver settings window. Once you have a first solution, use the Global Evaluation under Results->Derived Values to evaluate the expression lambda for the solution that you are interested in. Enter its value as the new eigenvalue linearization point, solve again, and repeat the process until the eigenvalue no longer changes. This should usually not take more than a few iterations.